 Welcome to the NPTEL course on remote sensing and GIS for rural development. This is week three lecture four. In this week, we have been looking at the Indian data archives that we can use for three specific indicators or data for rural development, which is water availability, soil health, soil data and climate data. We notice that there is less data stored on a single platform for climate and thereby we'll be using a different open source data archive, which is led by NASA. We will go through the steps of setting it up and how to find data, remote sensing data for rural development. First, NASA stands for because everyone is very famous this logo you would have seen on the earth and NASA going around. It's the space agency of US and it is stands for National Aeronautics and Space Administration. We need to be careful about using this widely in terms of what is the data for and reading about the data before explicitly using which is the metadata. I've already mentioned that reading the metadata is very, very important and is as important as using the data because if you don't know about the data, how do you use the data? So here also there are different ways to get the metadata. Let's look at some data archives. NASA has multiple data archives. And why did we come here? Because we could not get directly the rainfall, evapotranspiration, soil depth, moisture, snow cover, all these data. So we are here to extract those data for the rural development exercises. There are again tutorials on how to download the data. Please search for it. But before downloading, you should know how to search data. How do you understand the data, etc. So let's walk through today. We will look at NASA's Goddard Earth Science Data and Information Service Center, which is in short called GS-DISC. And this is a link for it. When I did my PhD, the link was different, so it does get updated. So make sure that you have always searched for GS-DISC, it will come. Or GLDIS data, it will come. This website is widely used because ISRO's data website mostly for Indian use. But then globally, if you look at globally which dataset is used more, or the data archive that is used more, it will be the NASA's. And this is based on publications and research profiles, portfolios, etc. On this website, it is also possible to download the data similar to the Indian system, and you will see a more robust visualization tool. So basically you will visualize the data on the website dashboard. And then if you want, you download it. Sometimes this is not available in other data archives. To run such a system, to run such a database, there is a requirement of high performance computing and infrastructures in the background. And that is what NASA has. Being a developed nation, the space agency NASA has a lot of budget to cater to the public's data requirement. And it is driven by public taxpayers' money, mostly the US taxpayers' money. So you don't have to pay anything to download the data. However, acknowledgements are greatly appreciated. So if you write your report, thesis, or paper, please do acknowledge that data was provided by NASA's data archive. So NASA has multiple data archives, not only GES, DIC. That is what I use widely, but there's also something called Earth data. And there's also called Geomani. And multiple more will come, but I'll just stop here. And it is, everything has different interfaces. It will be the same data. For example, I'm using Landsat, land use land cover data. If you go to each of these three links I set and search for that particular data, you'll get the same data. There's no difference. But the way of accessing the data, the way of visualizing the data is different. So here comes a good opportunity to test these different websites. We will start with one, and I'm sure that there's enough information that we can take from GES, GES, DIC webpage. It also acts like a dashboard because you are able to click and move sliders and those kind of things. So let us move ahead. And one more is Earth Explorer, which I've used also. You could see that all of that as .gov. So it's a government of US, .gov, .gov, .gov. In front of that, the domain or where the data stored can be analyzed. So here's NASA, here's earthdata.nasa, GSEFC, NASA, Giovanni, and Earth Explorer is USGS, US Geological Survey. Because most of these land use land cover, water is geological based survey, geology based science. So it is stored there also. All the data has its own metadata. So it's not duplicating the data in terms of efforts. It's just one server has the data and it gives to all of us. For example, USGS does not launch satellites. NASA does it. And it doesn't maintain the data, archive, database, etc. All of this is done by NASA. So here we should be careful that it is not both NASA and USGS working on it. It's mostly NASA, maybe USGS based for part of it. Okay. So as I said, we will look into the data for water resources and climate. Climate I will cover in the next class, but water resources because we will see more and more water data that is coming out. When you open GES DIC, you will come up to this kind of a page. In this page, you see a search box. So within a search, there's a search box. And here it gives an example. It says rainfall, GPM, TRMM data. So you can actually search for a parameter. If you go back to week two lectures, we have said that for the water sector, water focus in rural development, we need to understand the rainfall, discharge, soil moisture, storage, evapotranspiration, root, water holding capacity, all these things. Unless we have all the data, it is difficult to quantify the end product, which is storage. What is the change in water storage, which is going to be used for the future development scenarios. So in that case, we have a parameter like evapotranspiration, rainfall, soil moisture. You can put that here in that search box, or you can put the satellite name Landsat. Some of the Indian satellites are also kept here. And you will get a page like this that shows the different data. This is just rainfall. You could see that I have searched for rainfall as a parameter and global. So I will go through what these are in the live experiment. So the first box you see is like a calendar that gives you the number of dates, the days range in which you want to search the data. And then the paper like thing you see is a map where in the world you want to have it. So if you don't give these dates, all the data will come out. So we don't want that, but we will be careful in selecting India's part. And you will be amazed and interested to see that you will have better resolution data available for free in this portal. It also holds Sentinel, which is the European space data. It also holds some of the Indian database, but mostly it is NASA's database. So without further ado, let us search for that link. So we will be opening the web page for GES DIC. It's opening up. I hope you can see it. So when you click GES DIC, that link that I have shown in the slide, this page opens up. And this is the initial part. It says you do want to start the tour. Would you like us to take a tour? So I would highly recommend new students who are using it to start the tour. Since it's kind of not part of the course, I'll just close it and I'll give you the recommendations on how to use the data dashboard. Here, what you see is the preliminary things. You can also have a login. Without login, you cannot download the data. So I'm just going to log in. And I have already the login details set up. So you could see that my username and password also comes up. And I've logged in. It goes back to the front page where I initially was a high pen and it has my dashboard. You can have your own dashboard. Again, there are multiple videos to do it. But let's not get distracted with that. There's a lot of things that you can spend on this. This can itself be a full two to three lectures weeks. But I'm just going to spend two lectures on it. A lot of reading, a lot of papers, archives that have been put. Here is the archive size. So you could see that it is growing. So 3,000 terabytes of data is there. And you could see that now it's 0.211. So the 0.21 is MBs and GBs, et cetera. You will see that slowly it gets increased. Why is that? Is that in real time, data is downloaded from the satellite. It gets updated through the algorithms and being pushed into the system. Now you see it has increased, right? Archive data files. How many files are there? Again, this also has been increased. The files distributed has been increased. So these keep on ticking. You have 365 slowly. Before we even click within a minute, it will show an increase. Good. So 370. So now what I'm going to do is I'm going to click a parameter. Okay. Rainfall. You can do here like browse by category, subject. You have multiple subjects. If you want surface water for this rural development, we have surface water, surface air temperature, earth interactions, rainfall, precipitation is rainfall, land use, land cover, and then clouds, groundwater, and then glaciers, ice sheets, ecosystems, dynamics, public health, all these things. Okay. Or you can go as a measurement. So as I said, rainfall is one thing that we need to look at. It could be under precipitation or rainfall as R or precipitation. So precipitation amount, precipitation rate, et cetera. We can also take the vegetation, height vegetation index here, vegetation cover, land use, land cover, the water vapor, water flux, all these things. Evaporation, evapotranspiration is also part. The source you can go from satellites. These are all satellites. Look at how many satellites are there and sensors. So sometimes you'll see, for example, aqua. Aqua has air as MRC, aqua, modest. So all these are satellites and payloads. They have given different, different units. Yes. And then you do have some of the data products from different, different countries. Then processing level, high processing level, the projects, some of the projects are kept as aqua, for water, TRMM, for rainfall, and then GLDS, land driven models. GRACE is a program. So GRACE, DHM, for groundwater, et cetera. Temporal resolution, you can keep it as constant, like soil, moisture, soil type is a constant. But then you can look at how small you can get to per day, six years, once data, all these things. This is temporal resolution. So you can even get five minutes data, six minutes data. These are mostly for cloud and movement of clouds to study the cyclones and those kind of things. The spatial resolution is given as degrees or kilometers. I have said already one degree, one by one degree resolution is about 100 kilometers by 100 kilometers square. So the pixel size is 100 by 100 kilometers. You'll see the units change between kilometers and degrees. That's fine. You know the conversion, you can easily do it. So you can have as small as 0.9 kilometers or even lesser. Two kilometers you have. Submeter level also you have 10 meters, 20 meters, those kind of things. Features enable cloud, cover enable or not. So we'll just click this rainfall and then click. Just let it run. This first part will take some time because it is going to go through all the data sets and then pull out where it finds rainfall. I already had it ready, but I had to restart it. Let it run. But in the meantime, I'm also going to show you the other data searching tools inside GES. So it has 1, 2, 53 data sets associated with rainfall. You can store it as your favorite in your dashboard. And it says if you want more focus, you can say surface precipitation, infiltration, those things because it's too much. 53 data sets you're not going to look at and then see. It says hovering. So you don't have to download it. You can just click on this and then see. Not even click. Just hover and move your mouse on top of it. And then you will see how the precipitation goes globally. Look at the size of the pixel. It's all big. It is needed for high ocean currents, cyclone formations. For that, you don't get submeter or subkilometer resolution. You need bigger resolution. Good. So this is interesting. All the data has come. You can sort it by source. You can sort it by the version. All the time and resolution. For example, if you're a farmer, you need at least once a week. Once a day is OK. It's too much data. But once a week is also fine. So you can click on this resolution up or down to shift it to 30 minutes, which is very, very small. Very, very high resolution in terms of minutes. The most important thing here is also to see if the data is still current. If you see here, it has started in 2000 and 2020, and it's stopped. What does that mean? It means that the data, the satellite was processing the data, collecting the data, et cetera. But it has a lifetime. Most probably 20 years, 15 years after that they decommission it. They don't use the data because a lot of instruments satellite may drift. Friction, low losses might be high. So it's better to use it as much as you would use it in the earth. So normally a lifetime is 10 years, five years, but then they prolong it as and when the data is coming. But after that, after a particular time, they'll be functioning. Just think about it, all these satellites use solar panels for powering them. And most of the electronics and everything get deformed. So suddenly one day it doesn't wake up. So they are prepared, they are prepared and they will send the follow up mission. But still, like for example, grace was there. While the grace was collecting data, they send the grace follow on mission. So another mission of satellite which overlaps or replaces the current satellite. Okay, so here you could see all the data. So we are interested mostly in the 2023. You could see on the 11, it's already 10th today. So which means in some countries it is the 11th, 10th night here in India. So maybe Australia time it is 11. So it has the date for that particular part and all that. So it's 20 years of data, this 23 years of data, it's still going on. Okay, so now I'm going to kind of condense the data because in global and sometimes the global data is not good enough for all implications. You need to focus it on particular area. So let's take a data range. I'm going to take from 1920 you have data. So these are kind of hind casted, you have forecast data and hind casted, which means they know the satellite, they add some other data to predict rainfall. Now they use the same algorithm to go back. So I think 1960s, 1950s the satellite was launched one or two starting and so let's say 1990s the higher number of satellites came up, the technology was good. But they use that data and now you have like 30 years of data using the 30 years of data and algorithm, they could go back. So that's what they have done in some instances. Don't think that 1920 they were sending satellites to collect data. Okay, so you can click on it or just do this. Make sure you don't change the difference. And normally it's year, month and then date. Okay, so I'm going to use a disumber up to date. Okay. And then it says available date or you can choose from here also that is fine. And then you click this again. So the data has been set. You can refine it. But more importantly, when you click the map, the entire globe will come up. And we are not going to download the entire globe. Why? Because when we download the data and use it for India, it is not as usable. Right. You need to cut all the other regions and then use only for India. Okay. So these are the bounding box. You can actually bound a box or you can draw a box. I think drawing might be easier. So I'm just going to draw the box. Those who are having difficulties drawing the box, you can go as 68. Okay, 68 comma five is okay. You can put five. You can put 148. And then you could see that beautifully the India box has come up. You can also adjust this to 69 just to have more on the side. Okay. Let's say 166 is good. Okay. Now you've covered Gujarat 1000. So 66, 540. That gives you the box. Again, the boundaries are not as correct as the Indian government organization's boundary. But they have different boundaries. So don't worry about the international borders here. That is purely the data provider. Okay. And they have their own default. For example, the boundaries may not correlate with Indian boundaries. But when you download the data, the boundaries will not be there. It will be a box. So this box is what the data that you will be downloading. So I'm just going to click on that. Okay. The box is okay for me. And then click back the map and then search. So when you search now, it's going to be refined. You may not get all the 53 provided the data range. Yeah, the data. So for example, the first one will go off. There is no data for 2023 2022. So that will be thrown out. So now you could truncate what data you want. So while that is being loading, I also want to bring your focus to these left panel. Okay. So I've already talked about the spatial resolution, which is the pixel size. It has downloaded now. You could see that now out of 53, only 28 are within the range of this and intersecting your box. So that is good. We just keep that. You can mark it as favorite. And then you can come here and then do multiple other measurements. So I have put rainfall. Remember, so now you can click here. You can see what other things that you can monitor. So precipitation is 28. You can also get snow ice, topography, et cetera, et cetera. This is the subject. If you want to do the measurement, what type of measurements can you take? Okay. If you just type water, then you will take a lot of other measurements also like soil moisture, rainfalls also water. So you just say rain or precipitation rate. Let's say rain. Let's just look at the four data. And now only four data sets are there. Okay. So you see that model impact is there, aqua errors is there, which is very, very new. Just today's data also you can get. You can hover and you can see that the rainfall data, the IR precipitation estimate millimeters per day is given in blue, red and change. This is also the precipitation, precipitation. You could see that in India, it is raining in some regions. So you will see that particular location having rainfall. Okay. So it is not allowing me to use my mouse, but you can see it on the screen where the northern part, some slight rainfall is there and the rate is given at the bottom millimeters per hour. Okay. So you could see that and then you can see the full size image also here. You could see the three hours aggregated and the bounding boxes for this diagram. So you can see here like India location. So as I said, there's no boundaries. So don't worry about the wrong boundaries. The boundaries will not come into the picture. When they draw it, I think they get squished. So it's not actual boundaries for any country, not only for India. So don't worry about the boundaries here, but it's mostly the continents you will see. Okay. So you could then do a lot of data access and how to download the data. Every single equation, every single tab will give you the steps. Okay. And it all changes. So for that aspect, I will not be covering it, but here I have given you the link and how to take data from this website. So with the same bounding box, I'm just going to put water. Okay. And the same time. So let's see what we pull up. Okay. So now just putting water. Okay. You could see that 354 data sets are there for the data range and for the intersecting box. We will go to the measurement. As I said, just water, it goes everywhere. So root zone soil moisture is very important for me. I'll click root zone. And then you can also have soil temperature, soil moisture, water content. Okay. And then click somewhere else. Now it has come down from 354 to 31. Okay. And you could quickly look at the units. So the spatial is pretty big. Remember in the Indian soil database, it was 25 by 25 kilometers, kind of similar here. You have 0.1, 0.1, but 10 kilometers by 10 kilometers is still good. Okay. So there it was 25 kilometers, 0.025. No, I think 0.25. Yeah, 0.25 degrees. But now it is 0.01, which is 10 kilometer by 10 kilometer grid, which is approximately three times smaller than the Indian database, which is much, much higher resolution. And you can use these. The other thing it gives you is the different depths. So you can see here that the soil resolution, the timeline, do you want hourly? Every hour it takes a measurement or three hours. And then the project, which project you want to use, that also you can look at. Okay. I'm just going to take the root zone soil moisture hour. And I'm going to click this one, the first one, the second one. So what it says is it gives you three data sources. Okay. So spatial coverage is there for the entire globe. But mostly there are three different layers. It gives 0 to 10, 10 to 50, 50 to 100, like that depth, the depth at which it takes three to four cycles, it gives you. Okay. And that is important to understand the soil moisture at different depths. So maybe on the top, you'll see soil moisture, but it is not good enough. Why? Because maybe the farmer applied soil water, but it just stays on the top of the soil. It doesn't go as down. So you need to look at how much water has penetrated and gone in, for which you have a different data set. So someone might ask, sir, this is a satellite. It's way up in the space. How can it measure this water? So these are active satellites, which gives pulses. So there's a satellite, it sends a pulse, it goes into the depth of the soil and then gets reflected back. Okay. And that is why some, only some areas is being covered. So for example, look here, it's not the entire India being covered, only half of India is being covered. This is the evapotranspiration, but there are other data that you can take. Okay. So radar, data is more accurate for soil moisture. And then you have 35 data sets, as per the, you know, divisions and stuff. Yeah. So this is ground water, NOAA LSM, NOAA LSM per month, time series. Okay. So you have, you have details of how the data is driven and how they have been processing the data, what degree of solution and all those things. Good. So this is how you find data. Okay. And then downloading the data, accessing the data, there are multiple, multiple tutorials for this, like videos that you can watch. But the point is understanding the data is available for documenting the water is very important. The last one is evapotranspiration. For the same time series, for the same time box, the map of India I'm searching. And you can see that a resolution of 10 kilometers is there. And then you can also do 25 kilometers. Okay. Okay. Yes. So with this, I will just take one more minute to show you that there are multiple data sets. And you could come here down to see the data citation. As I said, when you use the data, it is good to cite those who have worked on the data because they are giving the data for free. For example, here Rodelstein has given it. Rodelstein is the hydrologist, chief hydrologist of NASA, who works very much in Indian water resources and stuff. You can see that you could use this paper for citations. And then you have documentation, the metadata. So you have a readme file. You can click, it opens up Notepad, a PDF document where it tells you how the data was reviewed. The same metadata, like how you access the data, what kind of processes were done, what is the resolution, all these things, how to read the plot, all these things. All these do get changed. So please understand that even though the data exists, the data format changes because they are keeping on adding the data and the database, they change the data formats. This is needed because too much data comes in and with advances in storing the data format, they will change the data format. So now it's in net CDF. In my time, it was just individual images, etc. So you can take it from online archive, you can click, you can go here in the online archive. I remember we said 2022, so we can click on 2022 and then download the months. So one to nine we have. So January, you can get the data and then download. So you can see here it is daily, 0102, so Jan 1, Jan 2, etc. So every month's data is kept. Then there is GeoMoney, Web Services, Data Source. GeoMoney is what I also gave the link for. That is also pretty much useful to collect the data for your images. And sometimes it does pull your analysis directly here because you're already given the date range, you're given the box. Everything is being pulled and here you could see. Let's do that again. You have the bounding box as you can draw the box, close it. And then the data range we said somewhere in 2022, 1st of December and then 2023, 2nd of Jan and then it automatically sees how many variables are there. Search, soil moisture, search. Yeah, it's coming. So you can see here a good resolution, the unit of the data and what depth, 0 to 10 centimeters, soil moisture. So as I said, 0 to 10 centimeters and then 10 to, there are four, five they have. 0 to 200 centimeters is also there. 0 to 10, 10 to 40 centimeters and then 40 to 100. So 40 centimeters to 100 centimeters, here you have 40 to 100. It's not sorted well, but you will find it. And then 100 to 200 centimeters or 0 to 200, 100 to 200, everything you have. So all these are aggregated together or in separate terms as you can see here, 0 to 10, then 40 to 10, 10 to 40, 40 to 100, 100 to 200, all these things. You can click on one particular data set and then download the data. So it will tell you like, okay, these are the layers that is done. These are the units. You can see that meters point is centimeters. So all these is 200 centimeters. So it's 0 to 10 centimeters, 10 to 40, 40 to 100 and 100 to 200 centimeters. So data above the data is given pretty well and then you can download the data here. You want to download? You can download or plot. Plotting is good because you can actually look at the data in real time. As I said, Giovanni is mostly for visualizing the data. You can have comparisons of two time periods or you can have a particular time series and then ask the computer to plot. You will have to have a good internet. It says some input, okay, start time has to be different. You can go back. You can sort it or just type sort in moisture. So for a rule zone moisture we have and then we're going to say time averaged or you want to time average between the data time series. What do you want? You can select here. So I'm going to say for a particular date I want a time series average differences. Okay, time average map is fine. Then you have to set a start date. So I'm going to say 2021 and then anytime is fine. And then you can say until 29 September they have. So let it do and plot. So now it's running in the background, the model. You can see the thing. So this is actually talking to the supercomputer and with NASA and using the NASA infrastructure to plot the map. Only when the map is plotted well and you see that there's no data gaps, you can download the data because most of the time you actually download the data and find out that it is not worth it because you actually don't have data there. Okay. Maybe we should have given a smaller bounding box. Yeah, let's do that. You can sort back to data selection. And here as I said, we will do the bounding box for India, this box, India I want and then go here. So it is loading the files and then tossing the input file. It should be faster than the doing for the whole globe. So now you understand that even for a supercomputer of that stature, it was very difficult. Okay. So here you could see that not all the data of the world is mapped. Only the India, the bounding box that we had is going to be mapped. Okay. Yeah. So you see here, these are the data that we use. The units, everything is given in the reading part. So you can do it. You can download as a geotip net. All these will directly go into the database of GIS. So all these have a geolocation already there, KMZ, PNG, net CDF, geotip. Okay. You can add subtitles. So for example, you don't want to download the data, but use this as an image. So geotip as an image, you can add titles, caption, legends, postline, United States, you can take them, you take the grid off, you can take the country's boundary. So now there's no issue of the boundaries. You can see it's clear, right? We're going to say, okay, the Indian region, the tile that we downloaded is going to be used. Okay. I think I've gone a little bit over time because I didn't want to break this tutorial. Sorry about it. But yeah, I think we will stop here. So this, I would conclude today's lecture. Please play with this a lot, the website. In the next class, I will again take you along with the GLDS website, NASA's data, so that you could at least look at climate variables. Now you've seen the water. Let's look at climate, precipitation, evapotranspiration, those kind of things. Thank you.