Downloading ESRI 2020 Global Land Cover Data

hello guys welcome to another tutorial in this tutorial i'm going to introduce you to a fairly new global land cover data set which was released collaboratively by s3 and the impact observatory institute now if you talk a bit about how this data set has been prepared it is basically making use of the existing sentinel 2 satellite image repository and as we know if we have multi-spectral images which are basically recording information that corresponds to different ranges of wavelengths we are able to derive the landcar by doing a classification and there are various types of image classification techniques as we also have discussed quite a few times on this channel in the recent past however in this particular case they have made use of a deep learning artificial intelligence classification model in order to derive these land cover maps using those satellite images of year 2020 that were captured by the sentinel-2 satellite and this global land car map has a resolution of 10 meters and the final product is basically categorized into 10 different land power classes and on the screen you can see what those land cover classes are so that's a bit about the background of the dataset now let's go ahead and see how we can access the data portal and then we will see how we can download some of these data for an area of our interest so you can simply head over to this livingatlas.arcgis.com lan cover website and over here you can find basically the general information regarding this product right over here if you click on this access map you will be directed to this page and as you can see over here we have a bit of a description about the product and what i'm interested over here is this map viewer so i'm going to open up this map viewer and right over here you can see an actual visualization of the classified global land cover map let's say if i were to zoom into some area over here you can very quickly get an idea what these different colors are actually attributed to by checking out this legend over here so basically blue color is attributed to water bodies and these red color patches are corresponding to built up areas and as you can see we have quite a spread of these yellow color areas which are attributed to the croplands and right over here we also have the presence of snow and ice along with some areas which we can presume to be some forested areas i believe so just like this you can see that the availability of this map basically extend extends throughout the entire globe and even you can make some changes to the appearance of this map simply by let's say increasing the transparency you can see the underlying map along with the labels of these major cities so that can be helpful for you to sort of narrow down your search and after that we can reduce the transparency of the global land cover map just like this so that you can start seeing the different classes again all right so this is how the data basically looks and now we'll see how we can download this data for an area of our interest so for that we i'm going to head back to this living atlas website and right over here you will see a link to download this data so i'm going to go over here and click on this link and that will open up a different web portal and if you would like to download the full global data set in geot format you can download it from here but that's going to be quite heavy 60 gigabytes and if you don't want to do that you can simply close this and after that you can see the presence of some tiles so what we can do is we can simply zoom into the area of our interest now for this particular example i'm going to download the data which corresponds to belgium and i have a shape file which demarcates the boundary of belgium and and later on i'm going to use that as a boundary to sort of clip the downloaded raster in order to get this s3 2020 lan data set for belgium and if i zoom in over here you can see that we have belgium right over here and that falls under this particular tile now it would be ideal if we could upload maybe a shapefile or a kmz file into this portal which which can show us exactly the boundary of of let's say the country or the region that i'm interested in so that i can make sure whether that area falls within let's say just only one tile or whether it's covering multiple tiles or not in that case we can decide to download the data which corresponds to multiple tiles but unfortunately i do not see an option to do that what we just have is a search panel over here from which we can search a specific area but but that doesn't really demarcate or show the spatial extent of let's say an administrative boundary or anything like that so in this case what i'm going to do is i'm going to click on the tile that i'm interested in downloading simply once like this and over here you can see an option to download that so i'm going to simply click over here and the process of downloading will get initiated and at the same time i would also like to point out the fact that this product is actually licensed under creative commons and if i click over here you can see that it's taking me back to the same portal that i was at before and if you check over here it states that this data set is available under creative commons by 4.0 license and if you're using this for any purpose it's good that if you can provide the appropriate credits where it's due in accordance with what has been stated over here so that's something to keep in mind and this is the raster that got downloaded so we can use a common gis software to view this raster you can use either qjs or arcgis software like that and i'm just going to go with arcgis desktop for this particular case but you guys can pick whichever the software that you would like to work with and then let's navigate to our working folder which is this one and right over here you can see that jotif raster so i'm going to simply drag this and drop it over here so this is how the data which corresponds to that particular tile that we downloaded looks and if we inspect some of the metadata simply by going to the properties we can see that it's fully geo-referenced and it's belonging to the utm zone 31 north so everything seems to be perfectly fine and now all i have to do is simply overlay a shape file of my area of interest in this case i'm going to overlay the entire map of of belgium which i'm going to pick from here and i'm going to drag this and drop it over here and now you can see that the data that we have basically is extending further beyond my area of interest so in that case i can confirm that i don't really need to download any other tiles and just a quick note if we need to perform a clip operation we need to make sure that both of these layers are in the same coordinate reference system but if i right click over here and go to properties of this belgium shape file you can see that the coordinate reference system is actually still in geographic coordinate reference system wgs1984 so we need to reproject this first into the same coordinate reference system as this one right here and then and after that we can perform the clip operation so since the coordinate reference system of my data frame which you can check by simply going over here to the layers and to the properties it's already been set to utm zone 31 north a simple way of basically reprojecting this particular layer will be by exporting it by going to data and export data and while doing that we can specify that we would like this exported version of this file to have the coordinate reference system of this data frame simply by clicking over here and i'm going to rename this to be belgium underscore utm zone 31 not shp it's going to be a shape file and we can click ok and we are going to add that now we can simply deactivate this layer and now if you check the coordinate reference system you can see that it has already taken the coordinate reference system of the data frame and now we can simply go to the search panel and say clip and we can select clip data management and my input raster is basically going to be this downloaded raster and the output extent will be defined by this reprojected shape file that we created just a few seconds ago and it's quite important that we tick this box use input features for clipping geometry so that this raster will get clipped exactly according to the boundary of this of the file and after that we can specify the output location i'm going to name this as lan cover belgium this be stands for belgium and and it's important to specify the file format we are going for a tiff format so i'm going to put dot tif and after that we can simply click ok and after the process is done if i simply go ahead and get rid of this uh unwanted layers so you can see that this is how the land cover distribution looks according to the classification by s3 and impact observatory institute so it's always good to cross-check the validity of this kind of a product because since it's using programmed algorithms to do this kind of classification there's still a chance that what the model interprets might not really tally or match with what's existing in that particular location in reality so it's always recommended that we do ground truthing as much as possible just to verify the accuracy of this kind of a data set but generally we can consider this to be a good product and if you are still wondering about the color scheme that we get when we dragged and dropped the original data over here if i open up the web portal that i was referring to at the beginning of the tutorial just to explore the data and if i set the two windows side by side in this kind of a manner you can see that it's basically still following the same color categorization that's being used in this original web portal so basically this area is referring to this region right over here and you can see that built up areas are represented in red color in both the images and the presence of trees in green color and the presence of water bodies in blue color and the crops in yellow color so you can still interpret the distribution of these different land components based on this particular color categorization and keeping that in mind you can use this data set for your further analysis so that basically concludes this tutorial i hope this tutorial was helpful for you guys if you do have any questions regarding what we discussed today in this tutorial do not hesitate to add a comment down below in the comments box and we'll get back to you as soon as possible and don't forget to subscribe to this channel as well if you would like to stay tuned for interesting gis related tutorials like this on a weekly basis so thanks all for watching guys i'll see you again in another tutorial

Channel: GeoDelta Labs

Views: 8,227

Rating: 5 out of 5

Keywords: land cover and land use in remote sensing, land cover mapping, land cover change, land cover mapping by arcgis, land cover classification of landsat images, land cover gis, land cover classification, land cover classification using sentinel-2 data, land cover change analysis, ESRi land cover map download, ESRI global land cover map, download land use data, download ESRI land cover data, ESRI land cover data, 2020 Global Land Use Data, land cover data, download land cover data

Id: gyJG1DktDNc

Length: 10min 51sec (651 seconds)

Published: Mon Jul 05 2021