 Hello everyone, welcome to remote sensing and GIS for rural development. This is week 8 lecture 4 NPTEL course. In this week, we have been looking at land use land cover as a metrics for assessing and improving rural development. Very focussely, we have looked at how accessibility to land and water are very very important criteria for rural development. On that note, it is imperative to have good data about land use land cover and how the land version happens from one form to the other, which is now documented through the LULC change. Land use land cover is one phenomena and then accessing between two different time periods or multiple time periods gives you a land use land cover change. So, in this lecture series, in the initial part we looked at the land use land cover, definitions, classifications, FAO notifications. Now we will be looking into the data that is available freely in open source systems for assessing land use land cover change. Please note that you also had looked into Google Earth as a very important tool, open source tool for downloading some metrics about land use land cover change. The last lecture, we had looked at the women's website and some data products. On the same note, we will continue the important tabs that are given in women's website. Even though there is limited data sets, both spatially and temporally, I am teaching women in one way to promote these software and data because sooner or later I foresee that this portal and ISRO's database will compete on international platforms. Even the budget available for this resource is much, much less compared to NASA or ESA. And that is where there is a lag of data and some delay in data processing. However, if you know how to do this through understanding the process and understanding the importance of the data, you would be able to do these maps by yourself. On that note, let us continue the discussion on using women's open source system. In the next lecture, I will touch base on the NASA system. I still stick a lot of my exercises, my mapathons with NASA data less used and a lot of new data used from ISRO and women because as an Indian, it is my duty to promote the Indian software and Indian space research data. How you can improve it is by contributing to these land use, land cover changes through events like the mapathon that is currently going on. Right now, we will come back to the exploration of this thematic service in women. And as indicated, we will open the first tab, which we already looked at UP as the highest number of rural population and villages. And then we looked at statistics, Indian state analysis will draw an area of interest. All these are the tabs. These are given in the tutorial, but it is not explained as a video tutorial as an NPTEL lecture. So I am doing it as a lecture. And then we will touch base on the statistics, analysis, metadata with WMS, as I said, I will not get in too much because that interfaces between GIS software and the women right now is looking at the LULC data and then we will do some overlay boundaries and data LULC opacity and degradation. So please allow me to share my web page. So this is where we start again, we can do a refresh to start from scratch when you start, you will have this India boundary and across India, even including Andaman and Nicobar islands, you'll see highlighted in yellow, luxury pilot, etc. So now, as I said, we'll be doing the land use land cover the latest version. As I said, 2015-2016, we will do UP because UP has highest villages and number of population, rural population and same view. So we stopped here where we analyzed the different datasets, different LULC timeframes we had, we had different resolutions, one is to 50, one is to 250 and one is to 10. One is to 10,000 is to focus on the rural urban periphery or mostly in the urban. So we're not going to use it. We can quickly show how it is done, say UP, not all states are there, you could see and only some districts are there. So these are the highest urban districts. And then we can just take this out for now. We can say view and you can see how the Lucknow city district has emerged. You can see the roads, the rails, drainage, some land use land cover, etc. So point of interest, I don't know, maybe it comes up or not, it doesn't come up so we can remove it. We can also say Tamil Nadu for Chennai, if it hasn't, there's no Tamil Nadu, but there's Puducherry and there's only Karakal, you can view it. And then you can see that this is Puducherry and there's no point of interest. So part of it is not coming up, so maybe they are working hard. This is the 1 to 10,000 K, very, very higher resolution compared to your other data that you have. I'm just saying, so this, see, Garigal, Amayar lake is a point of interest maybe, but it doesn't show on the data set, if you turn on and off, yeah, there is some things which come. So for example, this one year river, the bus stop is a point of interest. So if you click it, it goes up and down. So you can see that the river names college and then some smaller locations go up and down when we click. So this is important as the cartography image or cadastral maps, it gives you the boundaries of the urban settlements. There's very less agriculture here, so you don't see much stuff, but there is some plantation happening on the boundaries, some fallow land mangroves, et cetera. So let's go back to our initial discussion on UP, where we will look into the 2015 to 2016 land use land cover. Depending on your computer speed and internet speed, it will take some time. So please give it some time. And then you could see that this image has been populated. It's a beautiful image of the entire land use land cover in UP, with a lot of districts and village boundaries. If you zoom in, you will see all these boundaries. The big NH roads are also connected. So you can see that this, there is a small gray line which gives you the different associated boundaries. It could be district, village, tallow, boundary, et cetera. So as I said, let's go to Bayrelly. We have this because I want to also see the Ganges Plain, Plain and stuff. So you could see that there is a lot of rural settlements across Bayrelly, and you have this info tool already done. So what it says is you can also use this info tool. So this is really good to see what is the information on each person. For example, here it says as, the description is agriculture, crop land and the acreage is this, 255 hectares. Don't pay attention to the number of decimals, it's too many decimals. It's done because the pixels calculate by itself and you have different locations. So if you, for example, click on this one, it should pick up as a rural, built up rural, and it precedes a very, very, very small area. This is how much you can zoom in, you can zoom in as much as you can, and then it doesn't zoom in, it's here to move on. So this is a rural settlement housing, so you could see that it is built up rural. It's a very, very small one hectare, approximately the land holding size of the farmers. So now you have Bayrelly again, I will take off the info letter. So you have all this done for Bayrelly, the land use land cover is done, we are happy. Now we will go to the different aspects that are given in the table. So let's do the statistics for this particular location. So district wise for Bayrelly, so you can see that LUC information for 2015-16 for Bayrelly of the total area, you could see that 83% is agriculture crop land. So all the statistics you don't have to do. So normally how, as I said, a master's student or a PhD student doing the thesis would have to do all this by the GIS layers downloaded, give it colors, and then extract the pixels that are agriculture and then put a total. So the master's thesis will take approximately one year working on this GIS layer. But now with the click of a button, you get all this data for 2015-2016. So a quick question can come is find the, in UP, which is the district with highest number of agricultural land. And you can Google it and find it, yes, but let's do a quick Google and find that. But before that, we can also see from here. So for example, this is 83, 88%, but just looking at the different districts, you can see that other districts have very less urbanization compared to this Bayrelly. So Bayrelly, even though it's fully agricultural, there's little bit of a city, as I said, Bayrelly does take some part of urbanization. So that is 4.61% of rural area has been taken up, 2.45% is your urbanization, urbanization part. So if you look at it, what the statistics say is your built up urban is lesser than the built up rural. But in the image, you can see that red is big. So the idea here is, even though the red is big, the small, small rural urban, which is Maroon, okay? So the small, small rural built up accumulates into a bigger area than the urban built up. So this is where it's very important to do a land use land cover assessment because it gives you the accumulation of all these small parcels. All these small parcels is very, very difficult for the land surveyor to map to calculate the area and bring it as a table, whereas in GIS format, in using remote sensing satellite data, the color can be extracted and the color can now be cumulative added to get a net addition of total rural built up area. And now you could see that that is 4.61%. You could also see that the other parts is very less, 2.85% is the water. And that is this again, just tributaries that flow through by relief of these tributaries and the water bodies. But more importantly, you have a better idea now of the statistical division of this area. You can also do a statewide location. And here also you can select different districts. As I said, this is 86.4%, but the image doesn't change, so don't worry about it. But quickly we can see Agra has very less, it has more built, 74% of agriculture. You can have Aligarh, 85%, Allahabad, 75%, Ambedkar Nagar, 80%, Aurelia, 74%, this is certain random, Balampur, 72%, Visnor, 78%, so 80%, Chitrakot, 53%, Lucknow should be much less Muzah for Nagar, oh, so Muzah for Nagar has higher agricultural area compared to Bireli, and Barnasi would have more water resources, because that's where the Ganges flows. So you have 77% still on the banks of the river, which is an agricultural land. We'll have Lucknow, Lucknow, okay, there you go. So Lucknow has only 56%, approximately 12.83% is urbanized, which is very, very high compared to the other regions. And then there's agriculture crop land, and then fallow agricultural plantation, all these will combinedly be very less. So you could see here, how the land is land power has changed, so this is higher, Sarjan poor has 87.3% at this around 88%, you have agricultural activities happening. So this is where different, different statistics can be plotted up quickly for 2005 to 15, 2016, and it is giving you a total area of the district, and within the district, how it is divided. So if you add all these areas, it'll come to 4500 or 4575 square kilometers, if you add all the different land use land power types, that is, if you do a statewide statistics. You can do, you can see that, as per the statewide, 75% is agriculture, but Sarjan poor is above the average, right? It was 87%. So let's double check. Yes, 87%. So 75% is the overall state average, and now you have two averages, okay. So basically you can do this statewide, you can come down and then see the different land use land cover, you can see that 75% is your agricultural or allied agricultural activities. You can print this as an image, save as an image, as a PDF, and you can use it in your reports. Don't forget to cite, citations should be given to the data providers. So this is about the statistics. Let's do some analysis, okay, we're coming back to Bayreilly. And as I said, I don't want the entire Bayreilly, maybe I want along the river banks. So I'm going to say draw Aoi, I'm just going to click, and then you're allowed to click points. I hope you can see that I'm going to click points. And then once you do this, it will go like a polygon, I'm just going to go up here, then down. When you're finishing, just say double click, and it is done. So then you can click analyze, let it analyze your area of interest. So this is what I drew, now it has come up, and it will give you the statistics that you have selected 477 square kilometer, and in that cropland is the highest. You could see that very clearly cropland is the highest, followed by your water bodies, which is 10% point, river stream canal is 50 square kilometers. I purposely took the river canals because that is where the gang is tributaries and river is flowing. We could see where the agriculture happens, agriculture normally happens along the banks. So while I'm doing this, I also wanted to open the Google Earth Probe. So let me open the Google Earth Probe for you to analyze this Bairili region. We'll quickly see along the river where the Bairili is having agriculture. So the spelling was wrong, but now it just picks it up as Google does always. So you can see that Bairili is going to zoom in and the river channels are going to come up as and when the internet picks up speed, yes, you can see now. So these are the river channels. So now if you want to see back and forth, what are these? So you could see that this is what was labeled as wasteland, wetland in the land use land cover in the data. You could see these lands are waste wetland. But as I said, this is more fertile land because water comes along with water. There is alluvial sediments. So these lands are very, very pricey and have a lot of water for agriculture. And you can see a lot of agricultural activity happening. You can go back in time for this area. Always it's been agriculture. So maybe too much zooming cannot help, but 2002 should be good. This one part of the image is good. Let me go to 2005. Yes. So here you could see that not much meandering has happened. So you can see people farming right on the banks. So they start farming along the banks of the river because it is very, very fertile. You can see here, these are crops. The rose are giving the particular maybe paddy was grown on this side. There's nothing grown. You can also see how the tractors have laid the area. Both the sides there is good agricultural activity and beautifully the image captures all these aspects. Okay. Yes. So now I'm just going to go in a bit. You'll have better high-resolution temporal and spatial locations. You can see that this is being reflected in your initial areas. These are the areas where the pink color is being shown here as scrubland, gallead, ravineous, where the water is moving. These are mostly the land where it's kind of wasteland, not much activity can happen. But along the banks, you can see yellow color which represents agricultural activity. So you can do a quick analysis for area of interest or for a district. A district is fine, but sometimes you won't do only for a particular area of interest. You can zoom in and do, but make sure you understand the resolution limitations. Colorings, please don't use these kind of colorings. Just take, if I were you, I'll take these in an Excel. Just type these values in a table and then make your own pie charts. A black for agriculture doesn't look good, it should be green. So some coloring issues are there, but again, as I said, it is up to you. You are the user. You are going to present it to a committee. You are going to present it for research. So make it beautiful. Use an Excel sheet, use a table sheet, copy these values, plot it in a software. There's a lot of open source softwares that can plot these graphs. Then we'll go to metadata. Again, what is metadata? Metadata is the data about the data. So you remember that we had gone into the technical document. And then we read about this satellite, what was used, et cetera. So you can see that from this metadata also, which is needed for using in your reports. Who has done it? It is the Hyderabad Remote Sensing Center. The phone numbers, who you can contact. What type of data it is, vector data. So there's polygons, not rasters. So they've made polygons. Maybe they have converted the raster into polygon. And the polygon has converted, calculated the area. The resolutions are given. What is spheroid and datum, which is the geo-reference coordinate system. GCS we have used is WGS 1984. This is the same that we have used in our own tutorials also. And then the upper left, lower left, all these things. What data have they used? The original source is multi-temporal spectral data from Resource 2, which is List 3 Sensor. So List 3 Sensor was used. Some rectifications were done, which is cleaning up of the data. Most important, the source of the data also gives you the time they take the data. So they took Karif, Monson Season, August to October, post-Monson WD December, March, Re-Monson Zide, April to May. All these are given in the technical document also. But here you can quickly take it out. The metadata stamp is 5323. So where I did it, so today I'm doing it, which is 5323. So it is putting the date. And then you have the land is land cover type. English is the language used. Data identification, overall accuracy is 79% to 97% like water bodies. So what they're trying to say is the accuracy is OK, 80%, which is still good to do some good mappings and assessments in agricultural areas. Water bodies is very, very accurate, they say. It's around 97% accurate because water bodies are easier to map. Whereas agriculture, as you saw in the Google Earth image, there should be cloud cover. There will be some resolution issues. All these have to be taken care of, especially in the Monson time, when there is a lot of cloud cover. Because cloud cover will cover the agricultural crops. So you will not know what is under the cloud unless and otherwise you do a lot of survey, which is expensive. So where they did it, Meraz Ranchi, Corporate Name Lucknow and Brillard Institute of Technology, BITS, Remote Sensing Applications Center Lucknow, for the UP state and everything is there. So web services, as you see, you have a link to the WMS data. You can for QGIS, ArcGIS and other users, you can use this as a URL. So basically, when you put this URL in QGIS, it will pull the data into it. We are going to use much, much higher resolution data. So that is why we are not going to go through the QGIS exercise for this. But there is important overlay. So let's go to the overlay. So for this particular region, there are multiple data sets that you can overlay. What do you mean by overlay is there is a data and you put a data on top of it. So you're overlaying a data. So let's say, for example, we didn't know what was administrative layers. There's a lot of noise coming. So let's put a district boundary. OK, so I'm going to zoom out. Yes. So now you can see the district boundary coming up as red line. So Byri Lee has a district boundary. OK, so none I can do. So all that is gone. The state is already there, UP, so you don't have it. So Taluk. So within the district, there are multiple Taluks. So if you take this, it goes between two Taluks. So on one side, there is a Taluk which goes to the right side of the riverbank. And on the left side, there's another Taluk. OK, so this is the city Byri Lee. We'll keep it as the district. So the district boundary will catch Byri Lee. There you are. OK, so we can do none for now. And then we can take out the administrative layer. So just click the plus mark. It will open and then give you whatever choices you have. You can put roads, the national highways. So the national NH 24 is coming. You can see which is really important because when you're doing these rural connectivity maps, which is also needed for rural development, this is what you would need. You can remove the other layers that are creating noise. But you can create these national highway maps. OK, so the GQ is more smaller roads. Let me see if these come out and go. It's not coming up. So only some data has been uploaded. And the water bodies, you have reservoirs, rivers. So I'm just going to click on the river. The Ganges River should come up. So you can see the blue line just populating on top of it. So you see the light blue line. That is the Ganges and the tributaries. As I said, Byrony has the tributary of the Ganges. Since it's a lower order stream, it's not populating up. But we can remove it. So this light blue line will go if I remove it. And then there is reservoirs and lakes. So these are basically the small dams that are built across a particular region. So you can map that also. OK, so we can close this, close this. So these are the base layers. Now we have seen that there are thematic layers. Now we can see a change. So I'm just going to go to the AOI and then draw another AOI just for then analyze. It'll look to clear it. And there is no way to clear it unless otherwise we do work. Yeah, let me do one thing. I'm just going to zoom out, put an AOI here so that it picks it up. And then analyze, let it analyze, it won't analyze. Then we go to overlay and then we go to Byrony. So this is where we were initially. Byrony, we are here back again. You can do that again. Move, publish, view. We'll go back to Byrony. Second, it's easier to quickly do it by refreshing as long as you know the steps. So as I said, Byrony will go and then it goes. Okay, so we're back at Byrony and then we'll go to overlay. So now we've done the first base layers. We're going to thematic layers. Let's say 2005, 2006, how was it eroding? And we want to see UP. So here you don't see the entire district but the entire state will come up. And you could see that there is some erosion happening in the water body areas. And then you can, how do you visualize it is? You can see if I bring it down to zero, which is opacity. Does it overpopulate on top of it, et cetera? Okay, so the max you can go is nine and you don't see much change happening. Maybe here there is some erosion happening, et cetera. So this is only going to give you a erosion value if you have a land degradation happening. Example, let's go here. This is as much as I can go and I am going to reduce it. So here you can see that, let me bring my pointer. So here's there is some erosion happening and the color is given in the metadata of the erosion data. So you have to go back to erosion data, look at what the color means and then add it. So if you add, increase the opacity, you can see that this is populating up. If I decrease it, it'll go on. Same here, along the water body, there is land erosion and if you increase the opacity, which means it will be on top and it will block the bottom layer. Then you can see the maximum overlay happening. So now we can go back and then see if we could remove it. Yes, none. So we have seen the erosion. Now I'm going to show you the flood annual layers. There's only two, as I said, not all layers are mapped, only two are mapped. So we'll close that. Then the flood hazard also, asan is done. So let's close it. Land degradation, 2015. So this is important because this map is 2015. So let's do 2015 and then come back to UP. Okay, there it is. So you could see that these are the land degradation parts and let's see where is the landing. Let's say maybe it's saying that blue part is degrading land. Let's bring down the opacity and see where it is. Now, beautifully you can see there is a land use land cover on the bottom and on the top, there is the land degradation. Okay, so there is some kind of degradation happening, but you could clearly see that it is happening along the river channels. So if the rivers are not maintained properly, then there is a lot of erosion and degradation of land on both sides of the bank. So you could see here on both sides of the bank of the river, there is degradation. These are the high productive agricultural lands. So for agricultural development and rural empowerment, these lands have to be protected. So the water resources have to be managed. Okay, so I'm going to close this as none and then there's also land use land cover. So let's do a land use land cover 2005 and 2006. Let's go to UP and now you will be seeing a population of two datasets. Okay, so let's see if I slowly bring it down. You could see that this land was not in the agriculture. You can see that these are not agricultural, they have been converted to agriculture in 2015-16. So if I reduce the opacity, the top layer which is the 2005-2006. So 10 years before the base map, you could see that there is less agricultural land. So all these agricultural land have been wasteland, but because of science and technology, some interventions, all these have been converted to agricultural land. Where you see more change is the city itself. Let's go back and you can see that over here, just have an eye here. You'll see that the city is growing because I'm putting a 2005 land on top. So all these were agricultural land, the yellow, yellow, yellow, all these are agricultural land, rural land, but now if you open up the 2015-2016, you can see that it has been converted to the city, which is the bi-release city increase. Like this, you can keep on adding overlay, different data. So there's land is down to over 12, 2015-16, which is already here. There is salt affected and water logging, there's not much salt here. So you won't see that, but let's add water logging. That is pretty dominant, you can see here. So I'm just going to reduce the opacity. So this is the river and wherever the river floods, wherever the river expands, you will see water logging. So now you can download this map, you can print this map here, don't just download it, frame it, you'll get it up. I'll put none. And then you have the wasteland 2008-2009, let's do 2005-15-2016. It should normally overlap because we already have wasteland in this part. So you have this high wasteland around this area. So now you see that wasteland is already a land wasteland covered in the 2015-2016 data set, but it is not showing because it was not predominantly mapped in the 2015-16 data set. So this is it. These are mostly the overlay options. You have a spatial framework if you want, it's not much, it's just a gridding if you want. You can take it off. And then the boon data, if you don't want the boon data, you can take it out and then put rediff maps, which is the road maps, which is giving a higher number of attributes. You can see multiple roads here compared to the boon map. So you can use whichever data you would like to see. So it says the rediff layer may not be available at some higher zoom levels. You can take it off and visit. This is all about the ISRO LULC data set and how you can use it, play with it and download the data. Sometimes when you cannot download the data, you can always georeference it, print it, and then georeference it, and then use it for your study area. With this, I will close here. In the next lecture, we will look into the NASA data set for LULC and how they have done it differently, which is also important to share and copy. We want to see how different agencies are mapping so that we can have the best data available. With this, I stop here. I conclude today's lecture. I'll see you in the next lecture. Thank you.