 Thank you very much for being there, it's our partner in the EDU project and it's about Google Earth Engine and I will start with an introduction and then we will have a live demonstration and please remember that you have to choose the globe for interpretation in Turkish, you need to choose Russian. Please do ask any questions if you have any doubts. We are not so many so we can have this more interactive. As I was saying, we are now in our third webinar, we started with the methodological framework for ILU, this webinar is recorded, it wasn't live but it was recorded and it's uploaded on our webpage. Last time we had the fundamentals of GIS for land use planning webinar and today we are here in Google Earth Engine, next webinar will be about land productivity trends and land degradation and neutrality indicators and why it is important to merge these concepts in land use planning and use these indicators and our last webinar will be about digital soil mapping in our software. So what is Google Earth Engine, do you have experience with Google Earth Engine, if you do, you can please type yes or no in the chat so that I know it would be great. Google Earth Engine is a public data catalogue, it's not just one thing, it's many things, it's a public data catalogue, it's compute infrastructure for geospatial analysis, it also has an app and interactive app server and we will go through these different things during the presentation. Let's see in the chat if anyone could yes, okay, Mustafa said yes, good, so maybe you can also then share something about your experience and why you find it useful. We decided to include this in the webinars because it is a really powerful tool that wasn't available some years ago and I think it can be very useful for land use planning to obtain data, to process data and to visualize data. So regarding the data sets, what data does Google Earth Engine have? If you can please silence your microphone, Mustafa, I think you can talk about your microphone. Thank you. So it has petabytes, petabytes of Earth observation data and this is something in Google's mission of they want to organize data, we know that there is a lot of data available and previously you had to download it to your computer but nowadays you don't need to. Not only you have it accessible but also you don't need to download it. You can use it in the cloud and this is really a great advantage. So you have climate, weather, imagery, geophysical data here in this link, you can look at the catalog and find a lot of information of which other data sets are available. So I would say you have Landsat Sentinel models, digital elevation models, you have vector data weather and climate and the good thing also is that you can now upload your own data sets, vector and raster data. They have petabytes of information. It's really, really amazing. Not only imagery, you also have products and it's very easy to access it. You just search, for example, in this example, I put reflectance and you see which other data sets that you have available. For example, surface temperature from the thermal sensors, you have land and sea surface temperature products from several sensors. I think that for ELU project, which is also about being climate resilient, these data can be very useful because you have data for climate models that generate past and future predictions of different climate variables. So this is very interesting. If you want to see how climate, we're talking about climate change and being climate resilient, it's important to also explore what it is expected for the future, which are the different expectations and the history and the trends and what has been happening. Also, most data such as ozone data, maybe this is not that important for our project, but it's just another example. Regarding images, satellite images, well, you have all the Landsat products, all the Landsat images, you have all the Sentinel images, you have the MODIS images, daily, the 16-day composite, the derived products such as the vegetation indexes, NDVI, EV, snow cover, all of this is available just a click. Geophysical data, you have many products of land cover, such as Corine, for example, which is available for Turkey, the European Space Agency, global cover, terrain, many different digital elevation models such as the one that we were looking at last webinar, the SRTM data at 30 meters, also regional DEMs are available and national DEMs and many other geophysical variables, for example, there is cropland data, including cropland extents, dominance, watering sources, and this one that you can see in the image, for example, it's another example of another product that is nighttime lights that you can see, you can have a temporal history of the nighttime lights and this is, for example, a good proxy for urbanization, this could also be used. So Google is giving us a lot of storage and a lot of availability to data, which is not minor at all, but it also Google Earth Engine gives us the possibility to use Google's computational capacity. So we don't need to have a supercomputer anymore to analyze this big data or even pay or find a way to access cloud computing resources. We will be using, we will send a request and all the analysis will be performed in Google computational infrastructure, which is very high performance and great news for us. We don't need to have a supercomputer anymore. And we can do very intensive analysis. There are many possibilities to analyze data with machine learning algorithms and many others. And it's very fast. Things that would last many days now can last minutes or hours. This is a paper, it is a free paper. You can click here and you will see it here. You have much more information to read more about Google Earth Engine and this is also what we should cite every time we use Google Earth Engine. Planetary scale geospatial analysis for everyone. You see the title is exactly referred to this, though, that anyone can perform this geospatial analysis. So you can do, for the computation platform, you can do on-the-fly computation in which you send a request and you immediately visualize the result. Or you can also use batch computation in which if you have larger scale or if you want a specific scale and scope, you just send this and it is analyzed using parallel computation and then you will get the result. This is an example that is always shown of one of the things that were done using Google Earth Engine and before we had our maps for the whole world, our images were used with clouds because it is impossible to, we always have clouds. So using a simple algorithm and analyzing a stack of images, you can choose the pixels without clouds and make this world-free composite of the world. This is the map that is used as a base, for example, in Google Earth. This was called Pretty Earth because this is a map, an image of the world in which all the pixels are greenest, are the greenest they can be. It is always spring in the whole world at the same time in this map and there are no clouds. So it's not realistic, but it's very nice. With this, they did it for one year and then they did it for all the years with Landsat images and they produce this time-lapse video that you can access in this link and it's really nice. You can also access it through your mobile and you can zoom in, you can zoom out and you can see 35 years of Landsat images. For example, here we have a river. I don't know how it is seen in Zoom, I hope it's okay. But we can see how a river in Peru is changing in time and this is 35 years of satellite data, 15 million images of Landsat and Sentinel images and 10 quadrillion pixels. This is a lot of pixels and all of this could take a lot of time to analyze. But this was done in a few days using Google Earth Engine. So it is a really powerful tool and we can use it. I think that is the most important message of this webinar. We have this tool available for us and we need to take advantage of it. So this is a bit of what I'm trying to sum up what I was saying before then. You send a request from your computer and then you get the results. What happens in Google Earth Engine, well, there you have all the geospatial datasets. You have the storage and computation infrastructure of Google. So you send an analysis and it's processed there. In Google, Google does all the heavy work and how do you do that? Okay, there is a Havascript API. What is an API? It's an application programming interface. It's something that you need to translate to Google and you implemented it in the code editor of Google Earth Engine. You can also program with Python, not only in Havascript. It is a bit difficult if you do not have a programming skills or experience. But Google Earth Engine was designed specifically for scientists and to make an analysis for the environment and any applications in science. And it was not designed specifically for programmers. So it is very easy and you have a lot of tutorials and videos and materials to access and to learn. That is also what we want to show you today if you do not have experience. And this is the code editor. Here you write your code. Here you have different slots. You can visualize the results on the fly. For example, here I just searched for Corrine, you know, the land cover in Europe and it's also Whole Turkey. Here you can have your own scripts. Ouch, sorry. You can upload your own data in assets. Here in Inspector tab, you can query the layers on the map. You can click and you will get information. And you will see the results on the fly here. Okay. So, and this is the part that I find more fun of Google Earth Engine and where you can also be very creative and it's the applications, the apps. And because Google Earth Engine, and I think this is particularly useful also for land use planning and for the project, it gives us the possibility to create these apps which are very dynamic and anyone can access them just with a link. It's for experts and not experts. You just get a link and you click on it and anyone without a Google Earth Engine account, without any GIS knowledge, can visualize different layers, different maps, can also, you can program some basic functionalities, for example. I will show you an example right away. And these are, for example, some of the apps that have been curated. There are so many apps. For example, the Global Forest Change Explorer. Well, you can spend some time exploring these different applications. And this is what I was saying. I think this can be very useful, for example, during workshops with experts or during negotiation workshops with decision-makers and stakeholders to show the different possibilities, the different layers of information that we have. Also, to compare, for example, different scenarios. And as soon as I finish the presentation, I will go directly to this application that we did for the LDM project in Upper Sakaria Basin, where we put three different maps, the Corine Land Cover, the GSOC, the Global Soil Organic Carbon Map from FAO, and a Land Productivity Dynamic Map. So this is the Land Productivity Dynamic Map, the GSOC. And we put some functionalities in which, for example, you can draw a polygon and get the proportion of the different categories of the maps. But here you can be as creative as you want. And of course, all of these, there are so many people using these tools. And the basic idea of all of this is that it is free and you can share it very easily. You can share the code and this is great because you can take advantage of what others have been done. So you can use directly and adapt the code of others for anything, for analyzing data or for, for example, for the apps. So Cesar, if you would like to share the link in the chat, this link, if you click on it, you will be able to access this app, this application. IH Basin is in Upper Sakaria Basin. And to finish, this is where you can sign up for Google Earth Engine if you haven't already. It is free for education and for non-profit use. It is free. So thank you very much. I will now show you the, yeah, here as you can see, I put the link. So you create your app and you get a link. And then you just share it. And anyone, for example, can zoom in, can zoom out. Here, as I was saying before, we put three different layers. This is one that we are visualizing is Corrine LandCover, which was reclassified to seven LandCover classes, according to the UNCCD. And there is also this map of land productivity dynamics, where we can see here we have the reference, that if it is declining, if it is stable by stress, if the productivity is increasing, etc., you can zoom in, you can zoom out. You can have as a layer Google's map. My internet is very slow, so I shouldn't change so much. You can also have the satellite images as a base layer. And we also put the soil organic carbon map that was prepared by FAO. But you can upload the data. For example, for ISH, the idea is to create one of these apps using the information that we are developing for the area. We have a LandCover, we have many soil properties maps, soil map, etc., and also these productivity dynamics maps. And for example, I will leave it like this. What functionality did we put in this app? Well, you can, for example, draw a polygon, or a point, or a rectangle. And for example, here, I draw this polygon. I am a big patient and wait. And here on the right, a chart will appear in which we have the proportion of each LandCover class. For example, almost 69% of this polygon is cropland. 2.3 is artificial, 5.1 is tree covered, and 23% is grassland. You also get the proportion of the different land productivity trends. And the time series of NDVI, for example, in the last 18 years, these are annual values of NDVI. And of the GSOG, the mean, the sum, the sumatory, the maximum and the minimum of soil organic carbon. But as I was saying, you can be creative. For example, this is another app that we also created for Upper Sakaria. And I want to show you that you can use this functionality in which you can compare maps. For example, here we were comparing two different maps. Also here, if you click in a point, you get the time series of maps of NDVI. And here is the 16th daily. And this is the annual values of NDVI. So I'm looking at the chat. OK, so thank you very much for your attention. I will look now at if there are any questions. Also, now Cesar will show you in a live demonstration how to use Google Earth Engine. And we'll also go deeper on some of the details I gave. He's an specialist in Google Earth Engine. So how are we the questions, Cesar? I see you've been answered, Cesar. Yes, I've been answering all of them, except for this one. The last one that was recently translated. I can answer it now. The scale and resolution of these maps has simply said that there are many different products from many different satellites, and not only raw satellite images, but also products that have been processed by different institutions and people. And all of them have their own resolutions. So there are satellites with huge pixels, satellites with small pixels, and people do products on different scales. So what they have in Google Earth Engine is always the best available information for each of these products. So you will find things that are maybe like Sentinel. I will show now an example that is 10-meter pixel. But then you can have things like 50-kilometer pixel. What is the information available for your area? And if it is at a resolution, it will work for you. So some things are only for national level, and other things can be used and applied at the local scale. I will show later on the catalog where you can find a lot of this information about which is a resolution of the images and what sort of data you can find. It's a big catalog. OK, there is the other question. All of these images, satellite images, there are different type of satellites. Some passive satellites, which only receive what is naturally reflected from Earth. And there are also some active satellites that they are kind of like a radar. So they emit some energy pulse, and they recover it. In Google Earth Engine, you will find different satellite information. So there is information about the cover of the land that can be useful for geology, for land cover, for vegetation study, water. But there is also information in Google Earth Engine about a lot of atmospheric variables. So you will see the atmospheric column. And there are some satellites, some radar satellite that will show you what happens in the first, perhaps, layer of the soil. So there is a bit of everything in there. OK. Nobody has any question. You can still ask more questions while I move forward with the next part. And I will show you how to access. So as Ingrid showed, there is this Google Earth Engine web page where you only need to register and login. And once you register to have an account, which is free and open for any educational or also any government use in research or in this type of study. So you can basically very easily get an account. Let me share my screen. OK. So where is it? I will open the chat. I guess you are looking at my screen now. She put a nice, big yellow arrow so you can easily follow it. Let me know if the resolution is OK. So here is this CodeEarthEngineGoogle.com. This is the API. This is the interface. There are a few interfaces. As Ingrid showed, you can access the power of Google Earth Engine through Havascript to this web interface. You can also use a Python interface. And you can even connect it. You remember our last webinar on QGIS. You can connect QGIS to Google Earth Engine. So there are different ways to connect. But this is the most, for me, simple and most rich one. So what do you have here in this CodeEarthEngine? We have here several windows that do several things. Let me explain how this works. So first of all, you have this small window here to the left. Here is where you have your scripts, where you have your programs here. We will open some information that is around. There is another tab. This is the docs. This is the documentation. So here you will find the different tools and algorithms that are here with some explanation of what they do, what you need to input into them, what is the output. So every tool and every algorithm that you use will be described here. You can find a search. We will also try this in a moment. Here are your assets. So this is also a hard drive space where you can upload your information or the processes that you make in Google Earth Engine. You can store it here, or you can download it. But you can also have this space here. This is, let me show you the capacity. So you have here 250 gigabytes to store information that is produced in Earth Engine. Or you can upload your own data. So there is a lot of room and space to put a lot of data here. This is called assets. Hello, César. What do you mean by lanterns in there? OK. There is a question about the lantern index. I think there is an ingrid. Correct me, please. If I'm wrong, but there is a next webinar. You will talk about it. Lanterns in there are different indicator of LAN productivity trends. There is a paper. I think maybe, Emra, you referred to the application. It's in the application. Yes. Yeah, I just, when I opened the Sakaria Basin, so there isn't the, no, the engine. So there is a legend. There is a lantern index. It's NDPI, but it's multiplied by 10,000. So you have the annual mean of NDPI, but it's not in the scale of NDPI. It's not between minus 1 and plus 1. Is that why you ask the question? Yeah, so for the colleagues in there. It's multiplied by 10,000. OK, maybe just I can say that we can say that for the colleagues in the ministry. It's a vegetation index. So this data is showing the vegetation. So sometimes they are asking what is Corin, what is Lent Trend, Lent Productivity. So that's why I ask you. So, OK, it's a vegetation index somewhere. Yes, we will assess our set next webinar. It will be mostly about this, about how we can analyze productivity trends using vegetation indexes and map degradation. Yeah, thank you. OK, thank you. Good. So where I was here in the assets. So this is what you have here, your script, documentation and your files. Here is where it says in your script is where you write your program and a script. And here is where you have in the right side is the console. So when you output something, you print something, it will go to this window here. If it is a table, if it is text that you are getting out or a graph, but if it is a map that you are getting out, it will show here in the map below. So this is the map area. So how do we start working with this? OK, also here, very importantly, is a search bar. It says search places and data sets. So you can also start searching for data set here. So for example, if you are interested in Landsat images, you can write here Landsat. This way for my internet to respond, perhaps I will turn off my camera, see if that makes it faster. Ooh, taking some time. OK. Here it's coming. So here you have a lot of information. So when you search for data sets containing Landsat, you can see there are Landsat 2, Landsat 1, Landsat 8, Landsat 7. So you need to try to narrow down your search because these are a lot of collection of data that you can find here. If we could search, for example, Sentinel, Sentinel, we will probably find the Sentinel here, Sentinel 1, Sentinel 2, Sentinel 5. So for example, if I'm interested in Sentinel, which has really good images of 10 meter resolution, and I can come here, and this will appear. This will tell me about which is the data set that they have. So basically here they have all of the Sentinel images. Here is a description of what they have, which is the level of this area, of the level of correction, which is the type of layers that you have. In this level 2A Sentinel, because you have different levels of satellite images, this is the bands that this probe contains. So here you have a list of what are the different bands. I think I have the resolution of my screen very big. So you can see, but I cannot read all the information. But basically you have a lot of bands in this product. Some of them are 10 meter resolution, 20 meters, 30 meters, 60 meters. Here are some image properties. We will see also a bit about this. So every image the satellite has a lot of properties about when it was taken, in which angle the camera was looking, which day and hour, and a lot of information from the satellite that is here in the properties. And here is a collection snippet, which is how you can access this collection in here in the script. So it maybe seems a bit complicated at the beginning, but once you start, it's not so hard. Here in the script, here you can see I have owner. So these are my scripts. And then I have a script that other people share with me. And I can use for I have access to write. Other people share with me, and I can only read, but I cannot modify them. So this is very interesting for collaborative work. But here also very important, I have examples. Example give us a huge amount of example script for a lot of things that we can use and we can easily modify. So if you don't know maybe how to script too much, how to write a program from the beginning, here there is a lot of information. So for example, if I look for every data set that was here in the collection, so I was talking about Sentinel, I can come here to Copernicus 2, Sentinel 2, and it will open a script that will show me a simple function of this data set. So every land data set I want to see, I can just open the sample script and will show me how to load it, how to correct for the clouds. So there is a lot of information here. And Star Wars 2 does show me how I do to open Sentinel 2 image collections. So here I'm looking at Sentinel 2 image. We can see, we can zoom out and see what we can see here. So basically these are all the Sentinel images from 2018, the first half of year of 2018. And they are all being put a stick together. And they selected the average image for this period. So what I'm looking here, let me try to, okay, with my internet, we can move here to Turkey, for example, turn the image on. So I'm seeing all the average picture with 10 meter resolution for the whole world. So this is kind of the power of Google Earth Engine. I don't need to download, this will be a lot of terabytes of information. I can just open it and visualize it and export it. I will show a little bit of an example of how to do that. So here we are looking at this image, for example. On Copernico. Let me show you while this loads a little quick power point. It's only three slides, but just to make some concept clear. So if you remember last webinar, we talked about QGIS and different data models and we have vectors and raster. So what are the data models here? What are we looking in these data sets? So the names are a bit different and the way the data sets are organized sometimes too, but basically features are vector data. So all the vector data, they call it features and it can be point lines or polygons and they have a list of properties. So this is very usual as all vector information, but it also is different in the sense that usually when you have a shapefile, it can be a shapefile of points or a shapefile of lines, but not point lines and polygons in the same one. Here you can put everything. There is no limitation for the number of things that you can have and you can have different ones in the same. Data model. Another data model are the images. This will be the raster data model. In this case, with the raster data model, you can have one band or you can have many bands in one image. So you can have a single band or a multi band, but the difference with all the software that we normally have in a desktop machine is that here you not only can have different bands in one single raster, in one single image, but every image can have a different resolution. So this one maybe is 10 meter, this one is 60 meter, this one is 20 meter, this one is 10 meter. So there is not the requirement that you have in every software of having all the pixels the same size, the same projection. I can have different bands in different projections with mask. So the limitations that you have with a raster image in your normal GIS are disappearing in Google Earth Engine. And it also you have a list of property, a lot of properties about these images. And also another very important thing that you have here in these data types is not only image and features, which will be raster and vector, but you can have collections of images and collections of features. So a lot of images together, and there are a lot of tools that you can use to manage these huge data sets. So this will make sense in a little while. And I will try to show you in an example what I mean by this. Good. So let me move to an example here. I prepare here, I make like a new folder. This is called repository. So here you see I have a repository. This has a lot of properties. So if I come to these repositories, I can decide if I can share my scripts with someone. And if I make it for everyone to read, if I want to send writer permits so somebody can modify it. So this is very good to work in a collaborative way. I will open this script. Later I will share with you the link for this script so you can also have it as a base script. As I told you, I copy paste many of these things from the examples here. So there are a lot of examples. I will show you also some other examples, but this has to show you a little bit of how it works and the power it has. So here I basically click on my script and it appears here. What does the script do? First it goes to an image collection. So I open the collection Copernicus S2. This means that I'm opening every a collection that contains every image that was ever taken by Sentinel-2. So there is a huge collection of data. And then here I'm filtering. So I say, okay, there is a lot of data. I don't want to see all the images that Sentinel ever took. I only want to see image from 2020, from the beginning 2020 to now. Okay. And I also use this, this is a filter. So I say, I don't want images that are too cloudy. So I only want image with less than 20% cloud. So this is the language that is written here. This is a Havascript, a type of Havascript that you need to start learning. It's not so difficult as you see. And there are a lot of tools and examples that I will show and teach you. Then you can, here when you have a script, you can save, you can get a link for sharing, but most importantly, you can run it. Once you run it, you will get the result of this processing script. So here I see there is an image loading. Here it says Rb and this is an image. It says image, it has 16 bands. So there are 16 bands in this image. Each band has some information. And this is what I'm looking now. So what I did, I took every image from the beginning of 2020. And basically put it in this collection. It's called Datasets. So I created bar, Datasets equals to all these. So I took the whole Sentinel image, only took the first half of the year, only the ones that have little cloud and make a new collection called Datasets. Then with all these images, I calculated a median image of all of these and I just map out layer, just add it here for visualization. So this is what I did. I just say, okay, show me these bands. I can change what I'm seeing here. So if I want to put in an infrared, so this is red, green, blue. If I want to put in the red, maybe the infrared band. I can also put the infrared and change my visualization. So everything is available. So here I can put the infrared to have a better look at vegetation. And this is all 10 meter resolution images. So then I put this print mosaic. So this print will show me here in the console what this mosaic contains. So this mosaic is an image. And one of the question we can ask here, okay, how many images are here in this collection? So for that, I can quickly put this print Datasets and run it and see how many images are in this collection. So these are all the sentinel image from the beginning of 2020. And it sends me an error. It says, you have reached the maximum. So if more than 5,000 images, so they will not show me a list of every image because this will be too long and it will make my processor slow. So I say, okay, anyway, I wasn't that interested in knowing all the images from all the work. I'm most interested in working in IAS. So how do I start working in a specific place? One thing we can start by doing is making, if you remember the last time we were being here in a webinar was about Q-sheets and we have some IAS study area shapefile. So we have this shapefile of the study area. And one of the things that we can do is to upload a shapefile of this study area. So I can come here to asset and I can click here and say image afloat. I can upload my DM, any information I have. And I can also upload a shapefile. So I can come to shapefile. I can put a zip file. If I have a zip file of my shapefile, just one single file, or I can come here, for example, where it says, IAS study area, just select every file with the same name. Remember we say last time shapefile is not only one file, there are many files. So these are the shapefiles of my study area. It says already exist because I already did it to save some time. So I can put a different name and I just click upload. And here in the task, I will get it says uploading. It's already been uploaded and it will be soon, it will appear here in my list of assets. So I already have it here. How do I put my boundary of IAS in this map and I use it to make a small collection? So if I come here, I have different options. I can decide this boundary who I want to share it with, if I want to make it open for everyone to read, if I want to only allow some users to use it. So everything as you can see can be prepared accordingly. And here there is this arrow, which is pointing to my script. So if I press it, automatically this data will come into my script. So this is the IAS, I will name it IAS. And then I can add here, I already have it. As you can see, things that are with this double bar and turn green are part of the script that are not executed. So these are the comments. I can just put two bars and write anything I want and will not be executed. And if I remove these two bars, so now I add an extra line. I say, okay, I want images from the first half of the year. I want to filter by cloud, but also I want the ones that are around this IAS place. Now if I click run, let's see what happens. Okay, now I don't see the whole world. Now I only see this square here. So I have image only for this area. And as you can see my image collection has 26 elements. So what have I done here? I have gone from many thousands of images from all the world to making a collection of only 26 images. So I can come here to fit to and see which are these images. Here there are the 26 images that are in my collection. It started in 2020, January the second and the last one is from 2020, 08, 29. So there is an image from this last Saturday from two days ago already here in my collection. And here what I am looking is this medium. So this is kind of an average of all these 25 images. So if I want to see the last one, what, how was the thing last? Let's see how was everything last Saturday. So I can just change this and perhaps see, okay, from August 28th to August 30th Now if I press run again, I will see in my collection there is only one element. So there is only one image and this image will be the one I'm seeing and I know it comes from this Saturday. So this is the situation of two days ago for this basic. We can see here, if we zoom in, this is a 10 meter resolution image. So there's still a lot of zooming in to do. We can see some fields that are still having some crops. Some other fields that have been already tied. And mostly here in my collection, I'm going to show you some of the images and mostly here in the irrigated area, the streets. So this is what was the situation this Saturday. Let me refresh this and see if my study area too. So here is the one I was running my other task, loading the shape. So I have loaded two times, no worries. Okay, so here I'm looking at a sentinel of last Saturday. I can also here and comment this one, show my AS border just to know exactly where I am. As you can see, every time I click, this creeps zooms out and runs again. I have this function map center. This is the one that says zoom out in this coordinate with this level. I can just command the one so it doesn't move next time. And here I have the shape of my study area. So this was the IS basin project area. And this is the image of last Saturday. You here can do a lot of tasks. There is no much time left, I think. But here I have in this script that I will show you later. This is another way to commenting. So if I put a bar and an asterisk, it will comment everything until it finds an asterisk and a bar. So if I remove these two, all these lines that were inside will be available again. So what do I do here? I take this mosaic, which is the image that I have. As you can see, I have a data collection that now contains only one element and an image that contains only 16 bands. I can come here to my inspector. And this will allow me to have this crossair, which I can click and retrieve information for any pixel and show me, okay, which is happening in every of these bands. So this is the value for the 16 bands. And it also will show me information of what I'm impinging. We'll say, okay, this is an image. This is the bands. So I just make this. I took the band eight and the band four to make a normalized difference. Normalized difference, you know, there's the NDVI. This is a vegetation index, which is done as a normalized difference between the infrared and the red band. So band eight is infrared and band four is the red. How do I know that? I can come here to Sentinel. And here in band information, it will tell me, okay, the band four is the red band name four and band eight is the near 10 meter resolution each. So this is the two ones I use to run the NDVI. Also, if you want, you have any question of how to use this normal difference. You can copy here. The name come to docs. And here has I mentioned before, there will be a information about everything that here it turns this color kind of a violet color. You will find information here. So if I click, it says normal difference. This is a tool that compute the normal difference between two bands. So you need to give this two bands and it returns an image. So what I do here, I take my mosaic. I do a normal difference. I clip. Clip. Remember from QCC. It's cutting the image. I clip it with the highest border. And I call this NDVI. And then I say, okay, let's show NDVI. I put a minimum and a maximum and I decide which colors I want to see on which name. So next time I run this script now that I have made this to available. It will open the image. It will open. Hi-ash polygon. And it will open the NDVI. So I can turn here. I can turn the image down and I can see now only the NDVI of hi-ash. And I can see here what I was looking before. There were these areas. They are still very productive. I still have high NDVI. I can click here now, for example, and say, okay, this is the information on all the 16 bands. And in NDVI, it's 0.79, almost 0.80. So it's a highly productive place at the moment. And here, for example, where it's red, it's 0.09. So perhaps here where I click, I don't know where. It's a bit of a bird soil. Or very little vegetation at the moment. So as you can see, you can also compare the actual visible layers with the NDVI. And you have this information. Here I also commented another thing. So make another mosaic from April. So we may be interested to say, okay, this is the situation this Saturday. But how was this situation on April? So I will just uncomment this. And take this out. So this whole section of the script is uncommon. What I did here, I basically copy the same. It was written before. So I have another instead of dataset. Now it's called dataset two. I open the Sentinel, but instead of having here, August to see the last one, I put from the beginning to the end of April. Then I did also the mosaic and I calculate the NDVI. So here I will give me another called NDVI two. I can even rename it and put NDVI April, for example. And if I run now this script again, I will have here my image, I-ash boundary. I will just leave only the NDVI. So this is my NDVI from August. And the here on top is my NDVI of April. So as we can see here, I can also make this layer larger. If I press this icon here, I can make it the whole window. So as we can see in April, there were a lot of area that were cultivated around here at the bottom of the, of the basin, but then in August, most of it already been already collected. And it's now holding no vegetation. So we can see this here, or we can also zoom here and see the images and see, okay, how was the situation? This is what, what was the situation for this place in April. And this is how I was in, sorry. And this is how it's on Saturday. So you can compare what happens from April and what happened on Saturday, which plot. So for example, there are some fields here that were with no vegetation on April, but now on last Saturday, they have already some vegetation. So, okay, just to show an example of something that has good resolution and it's very new on time. So one of the question is how do I take this information out once I process? So as you can see, I did a lot of stuff because I took a lot of information and just process. And now I say, okay, I did all my process. I want to get it out to a raster file. This is the last section, download any of this data. So if I just uncomment this and open this section, it says export image to drive. So how do I get these results out? I can either store it if I want in my assets or I can export to my Google drive. So I say, okay, export the NDVI, call it NDVI IS August 2020, 10 meter resolution for the region of IS and put it in this folder. So now when I run my script one more time, it will do everything again. Get the image, calculate the NDVI. But this time it will appear here in task, this option to run this task. If I run this task, it will say, okay, there is this task with this name. I'm already did it, so I will put another name. 10 meter pixels to Google Drive and call the file NDVI ISO. So if I run this, I will send a request to Google servers to process this information and send me the final raster file to my Google, to my Google Drive. So here I come, for example, this is my Google Drive. It automatically creates a folder called IR export and inside it puts this layer, which is 60 megabytes. So something that maybe it's a lot of gigabytes of information, I process everything, as you can see very quickly. And then I get here in my folder, the IS NDVI of the basin. This I can download, which I already did. So we can come here to the folder. So I already downloaded this file. This here is a 60 megabyte file of AS and I can just open in the project we were doing before about NDVI. I can just open this file here. So now I have, I still have the DEMs below. Now I bring broke here this NDVI. You can export all the bands, any bands you like or any image, but this will be the NDVI of last Saturday, two days ago for my study area. So here you can see this NDVI. I don't know how are we going with time, Ingrid? We are okay, but I think it would be good to see if anyone has any questions or any comments. Now we will show you how to access to all the educational and training materials that Google Earth Engine has on the web. So the idea of today's webinar is just to give a glimpse of what can be done with Google Earth Engine and how useful it can be. Not only for land use planning, but for many projects. And if you want to train yourself or start from zero, you can do it on your own. So now Cesar will show you that, but I think it's very nice if you have any comments or questions, or if you want to share previous experiences with Google Earth Engine. Mustafa, if you want to turn your Microsoft on and we have the translation. Okay, so what I show here is... I am sorry, I just wanted to thank. It is so nice. I mean, such a nice presentation. I mean, in the last five years, I haven't been engaged with remote sensing, etc. You have taken me to the past phase. Actually, I really liked it. And we have seen that we can have access to data very easily. Thanks for your presentation. Thank you. Well, actually, how can we have access to the materials, guidance, or any materials which can guide us to help, which can help us use this engine more? Actually, if there are materials out there, can you kind of share those materials with us? It can be a booklet and a thing so that we can do further reading and we can learn about how to use it more detail. Thank you very much, both colleagues. Thank you. Both presentations were excellent, by the way. Thank you. Yes, we will share the presentation and Cesar will share the code also if you want to start playing. And now Cesar will show you how to browse in the internet. But if you just, in the presentation I sent for each part, there were many links. If you click and you start exploring on the web, you will see how easy it is. Google is so user-friendly. I think it's foolproof. If I understand it, I think you will find a lot of material to explore by yourself. So now, Cesar, if you want to show that, we strongly encourage you to start looking by yourself. Good. Yes, so here, for example, if I want to share with you this script, I can get here, get the link. I will copy and I will pass it here. Remember, in order to open it, I just put it in the chat box, but in order to open it, you will need to already register before and have the Google account already registered for Google Edge. It's very easy to do. And regarding the examples, as I showed you also before, here there are a lot of example scripts. I will show you some others in a minute, but also a very important place for this Google platform is here this question here in the right. This question mark has the user guide, shortcuts, a lot of information. Here in the user guide, if you click this one, a new window will open and it will tell you, okay, get started. It will take you to where to go. It will explain you how to open your account. Very importantly, you have guides. For me, this is very useful. Here in guides, you have this information that you can start reading. If you are using also, I don't know which other language they have here. If it's Turkish available, but you can also translate it with Google Chrome. And there are here a lot of information about how to start working, how to start Google Edge engine. Most of the basic things we show, but also information, okay, images. What are the images? How are the images in Google Edge engine? You have codes, examples. For example, I don't know, image visualization. You don't remember how to do that. Here is an example, opening Landsat and changing which bands you want to see. You can basically copy and copy paste this in your in your script. For example, if I come here, I have an empty window. I can pass it here and run. And I can run the same exercise that was here in this example. And I can touch it. I can delete this, put a different number, play around. For me, these guides are very important. It has a lot of information about how to manage image, how to manage image collection, features, and a lot of examples. But also, besides these guides that are really good, you can come here what says community. Here in community, there are also two different sets. The community tutorials, which are also amazing. So there are different, for example, forest cover and loss estimation. So here is a tutorial. You can see the code. They share all the information and they show you how to run a process. For example, here to analyze change in three land covers. So all the script is here, all the explanation. And you have different different type of exercises about different things. So here is, for example, how to make time series animation, a temperature of land surface in Uganda. And everything as I show you, maybe this is in Uganda, but you can change, you can put your own area and already have all the script prepared for Uganda, but with a simple changing, your shape file, you can go directly to IH, for example. So that's one of the great ventures of Google Earth Ancient. Here I'm here in San Francisco. So I'm looking at this image, but if I can change the coordinate, I can do the same exercise as I showed you before in Turkey. Also here in educational resources, so remember we have this, I just came to community, there are the community tutorials, but in educational resources, you have two things. Edu, we contain, these are different courses on Google Earth Ancient. This is by the University of San Francisco. So this is also a very nice introductory course. So you can just click here and you will access this free course, which also contains explanations, the codes and everything that you can run this exercise. There is also another exercise here from the San Diego University, from Yale University, Pennsylvania University, some different courses that have been teach in different universities of institutes, aiming at different processes. I think this is more about forestry. And also there are also in some language, I see some Russian, but there are a lot of material here. And also in training resources, there are some other information for trainers and some other exercises. So here there are some, these from the Cervir in the Mekong, they do a lot of things with the trends in vegetation. So there is a lot of different, this is one from analysis of precipitation, temperature from Columbia University. So, well, you can look at yourself. I think I've already went through all of them. So a lot of information here for self-learning. So as Ingrid said, you can follow the links that are in the PowerPoint. Or once you come here to this code editor, remember here in this question mark, you can always come to the user guide. And here you will have the guides. And in the community, you will have a lot of tutorials. So there is a lot of documentation. And also here in, you have, remember examples of how to do the normalized difference, how to do hillshade, how to clean and uncover. If you want, there is also much more demos classification. So if you want to know how to classify, this is a very easy example on classifying land cover. Let me show you. So this is a very simple script. You basically create some points and put some points on the different land covers. And then you use this cart methodology, which is a very, very interesting methodology. You get the confusion matrix. And you can classify the land cover. Here is an example. But as I say, you can just copy paste from here and put it in your own code. So it takes a time to get used. But there are examples for many different things that you can use. I mean, this is also what we saw. Remember last time in QGIS when we worked with the DEM. Here is opening a DEM and it's doing what we did before, just calculate the slope, the hillshade, put some style to this hillshade. And I think this is done for the whole world. So you can come here, you can put your study area, you say, okay, this is my study area, and you can export it. If you don't know how to export it, you come here to my script. This part, as I say, export, you can just copy it here. And instead of exporting NDVI, you can export this other layer. You just change the name. Instead of IS region, it will be this region, this geometry that I made. So I write here geometry. So this is much explained in all the tutorials. And as you can see, just with few changes, cutting and pasting, I can take a lot of information and process a lot of data. I don't know if there is any more questions. I don't want to bore you with showing you stuff. There is a lot to show. Maybe you can show how to get an account, because you need to have an account. You have to be a newser and you need to be approved by them. And also the data catalog, I think that could also be interesting. Okay. You can come here to Google Earth Engine. So you just look Google Earth Engine in Google and you can come to the main page. And here it will say, okay, datasets. Here you can also access the code editor. It's the one we were using to run. But the first thing you need to do is here we'll say sign up. So you need to register for an account. So just look for Google Earth Engine and go to sign up. So here at the end, it's also this, sign up now and it will show you how to register. So if I click, it will say already, I'm already signed up. So there is no point. I show you much here. Here you can also come to datasets. This is where all the dataset, the catalog is. So as you can see, there are some weather information from different models. There are some, there are the Landsat images, the Sentinel images, land cover information. So there is a lot of, here are some overview, but you can search depending on what you like or you can put view all datasets. Maybe this takes a little bit. Okay, so here there is all the datasets is a lot. So here you can see, it's huge. And you can filter. So if you're looking, I don't know for fire, fire products. Okay, you can see here different fire products. This is the fire films, fire information management system. Here gives me some information. I can click on this catalog and it will take me to the page where it gives me more information. I can click on the Internet. It's a bit slow. And this is also the Modifier products. I can click, okay, here I go. So for every, every this, it says how you can, what is the name? This is available from 2001 to 2019, for example. And it gives you all the information. And here it gives you a piece of code. So if you want to test and you want to see, you can just copy this, go back to your code editor, paste it and run it. And it will show you this fire product. You later, you can change the date. I mean, this is for 2019. Perhaps you can change the date to another one. It takes a while to run. Okay. Let's see if there is another one. But basically you can get this, you get the idea. So you can, this is the films. This was the one that I was opening before. So this is from 2000 to 2000. So to three or four days ago, there is images. Also here you have a code. So you can just also try it, reserve, clear script, run it. And then we'll show you some of the, of the fires of 2018, August. You can change this. Of course, you can put 2020. Yeah. You get the idea. And easily modify the script to, to show information in your area. And if you want to download it, you can download it to Google drive. So you see this is available for the whole world. And if you want, you can change the location and the data. I don't know if there is any more question. I think one of the big lessons from Woodruff engine is how important it is to, to work in an open way and share information that is the, the big power of this platform. And having all the information at one place and very accessible and all the people working, sharing their codes and products. If you go to also in the, in the page and their case studies and many papers and products that people are working with and designing and thinking on how to produce a specific information and then they share the code and you can use it. And, and well, this is like the new paradigm, I think. And it's important to, to know that these tools exist and start working in this direction. I think it would be more powerful and useful for all and to reach our objectives, which are sustainable sustainability in general. So if there aren't any more questions, I would like to thank you all for participating very much. It's nice. Invite you to the next webinar, which will be about land productivity trends. And thank the interpreters very much and the organisators. And that's it from me. Thank you very much. Thank you, Denise. If you yell, thank you very much. There you are. It's nice to see your faces. Thank you. Thank you. Thank you. Thank you. And we will share all the info. And if you have any questions, please, you can ask us by email. Or next time. And see you soon. It's in two weeks. Not next Monday. The other one. Thank you, everyone. Thank you. Thank you very much. Bye. Bye bye.