 Thank you very much for being there, it's a pleasure. Today we have our fourth webinar on this series of webinars. The topic for today is planning for neutrality, considering land degradation neutrality in areas facing integrated land use planning. So as I just said, this is our fourth webinar. We started with the methodological framework for INU, which was recorded and uploaded in our webpage. Then we had a webinar on GIS for land use planning and demonstration using QGIS. Last time we had the introduction to Google Earth Engine and today we will focus on some indicators used in land degradation neutrality and considering this framework in Edo. Next time we will meet for some digital soil mapping with our software. Oh, there is an E missing there, sorry for that. So we are about to finish with this series of webinars. So let's start with today's planned topics. We will start with an introduction to land degradation neutrality and its link to ILU. Then we will see more in detail, which are these three indicators, change of state indicators. Then Cesar Garcia will give a demonstration using trends.Earth, which is a tool, a QGIS plugin to calculate these indicators globally. Then we will focus more on land productivity dynamics and trends of vegetation productivity as a very key indicator to map land degradation that can be used in land use planning. I see the chat is okay, Eili, should I stop to give a place or I continue? They couldn't form the problem, you can continue. Okay, you just let me know if I should stop. Okay, so I don't know how many of you are familiar with land degradation neutrality. This concept is defined as a state where the amount and quality of land resources necessary to support ecosystem functions and services remain stable or increase. This is why it is called neutrality because they either remain stable or even better they increase, but they do not decrease within particular temperance spatial scales. And LDN was included as a target for the sustainable development goals which is life on land and it's target number three. So SDG 15.3 is the target that refers to land degradation neutrality. So we want to achieve land degradation neutrality by 2030. Okay, and why is it important to link land degradation neutrality to land use planning? What are we talking about this in this webinar? Well, the neutrality mechanism, which is this that I try to show here is based on counterbalancing future losses that we can anticipate and future gains. And to achieve this land use planning is key. It's basic because if you can anticipate when you are doing land use planning which are the places and areas in which you will lose your natural capital where areas will be degraded but you take this into account to anticipate possible gains and then you are doing a lot for achieving land degradation neutrality. So it's really important that land use planners are aware of this and try to consider it because if not, it will be very difficult to achieve land degradation neutrality. So at which scale this neutrality mechanism needs to be thought of and achieved? Well, not in general, but in specific land types. So within unique land types. And what do we mean by this? Well, land types are defined as a class. So you classify your area and you take into account different variables. So it's a combination of edific, geomorphological, topographic, hydrological, biological, and climate features. This is of course an exercise that is not easy to do. But for example, for Irish we still, we have a lot of information available that we could use to define these land use types. So here I'm taking this map as an example of a land use type actually is a land cover map. It's not derived from all of these variables, but we could think of it as an example of land use types. And let's have a very simple example to see what we mean by this neutrality mechanism. So let's think for example, of grasslands as a land use type. Now here in this map, these are grasslands, like they are maybe a bit different to distinguish. They are not like violet, but a bit brown. And let's imagine that when we are during the land use planning process, we can predict, we can anticipate that 300 hectares will be converted from natural grasslands to agriculture. So what we need to do is to think of a way to counterbalancing these laws because a conversion from a natural grassland with more biodiversity can be considered a loss. So we can anticipate 300 hectares of loss. And we need to think of a way to counterbalance this again. And for example, we can decide that in some areas, for example in this, this is a very illustrative example. Now of course, in these 500 hectares, sustainable grazing management will be introduced. And this would be a proposed game to counterbalance the loss. So this is basically what we mean by neutrality mechanism so that no net loss is achieved. However, it is important to also consider the magnitudes of these losses and gains because it's not the same to, for example, deforested, you cannot compensate deforestation by sustainable, but by, for example, sustainable forest management because in maybe one thing has more magnitude than the other. So magnitude has also needs to be considered. Anyway, in this example, for example, the direction would be okay because we are counterbalancing gains with losses. This is a whole issue that needs to be discussed, of course, and considered for each area and each project. And sometimes you do what you can. The important thing is to have this framework in our minds when we are doing land use planning so that we can anticipate and plan for not losing natural capital. So this is what we have been just talking about. This figure, if you are familiar with land degradation neutrality is like the key here to summarize what land degradation neutrality is. It's in the scientific framework for LDN. And this is the part that refers to the neutrality mechanism. As you can see, this is on top of this pyramid which is the LDN response hierarchy. What do we mean by this? This is related to what I was just talking about. LDN is based on the fact that it's better to prevent than to cure. So it's always better to prevent and avoid land degradation. So these should be the pillar and the efforts should be here in avoiding land degradation. The second part is reducing and this is related to sustainable land management practices. In some areas we can introduce sustainable land management practices, sustainable forest management, sustainable grazing management and this way we can reduce the degradation because the thing here is that we know that we will degrade land. We cannot, sometimes we cannot avoid it and we need to counterbalance it by, for example, reducing the degradation. And last, we have reverse. And this is related to restoration and rehabilitation of land. And before reversing, this is very costly and sometimes very difficult and actually it's almost impossible to restore an area to its initial state. If we are considering biodiversity ecosystem, other ecosystem services, et cetera. So this is in the top of the pyramid of this response here. So how do we do this? Sorry. How do we integrate LDN in land use planning? Well, there is no simple answer for this. I mean, actually I wanted to show you that now this competition is open and it's been organized by the GEO LDN Initiative and the UNCCD and they are asking for teams to compete and to present ideas on actually how we can design a solution for land degradation and alternative using land use plan to make tools that can be integrated to these trends.Earth software I was talking about to help decision makers to include land degradation neutrality in E-loop. So this would be very interesting if any of you would like to enter the competition. It is still open until the end of September. There is a lot of information and ideas are being welcomed by the UNCCD, for example. The neutrality makers base anticipates and plan in the base land and the monitoring indicators. Well, we need a way to monitor what is going on and to report and three indicators that were used for the UNCCD reporting were also chosen for the SBG. I mixed the letters, SGD, and these are very three global change update indicators. Which are land cover change, land productivity dynamics and changes in soil organic carbon. These indicators are not additive and they are related to each other. And if we can map these things and maybe we can use this also in land use planning. So let's see how it is calculated, how land cover change is included in this framework and how it is recommended to calculate this indicator with default data and methodology. Of course, each country then can tailor these to their realities and available information. We will talk about this later. So the default data for analyzing land cover and land use changes. This is a key indicator because land use changes of course affect degradation. We talked about this a few slides before and the default data is the European Space Agency data on land cover. The good thing about this data is that you have land cover maps for many years. So to see land cover changes you need to have two land cover maps to be able to compare them. And these two maps needs to be comparable to. For example, for IAS we have a great land cover but it's only for one moment. So we would need another land cover to see changes in time. This is the default data. So it has many categories, this map, these products and these need to be regrouped into the seven UNCCD categories. So first you need to regroup these categories and create a land cover map. For example, for your base year, which is 2001. And then for 2018, for example, you do the same and then you see what happened, what things change. As you can see before IAS at this scale this data set which has a 300 meter resolution is not very informative. So basically I didn't put now when we do the exercise we can see with more detail these maps. I didn't put the references but it's mostly arable land, cultivated land cropland and some grasslands using the ESA maps. Once you compare these two maps you obtain a land cover transitions map. So as you can see there are almost no transitions in IAS between these two years using this data. Only here we can see some grassland loss or cropland loss and then you analyze this and you obtain this is all using drens.erf. So then you have these metrics where you can see which land cover types changed how much, which are the areas that changed for example from grassland to cropland, et cetera. And then to obtain a degradation map using this indicator you need to define which of these land cover transitions are positive and are negative. So for example, and this is what this matrix is about. For example, here converting a grassland to cropland is seen as a positive change but you can actually change this. It's very easy to change it Cesar will show you in drens.erf how to change this or you can use your own data sets. If you have land covers, for example, for Turkey we have Korean land cover also which has many years and better resolution we could use that or any other land cover that you consider more appropriate. But you need to define for your area which transitions are positive and which transitions are negative. In general, going from forest to any other land cover type is negative, for example. Then so then you change this map into, from this map you can obtain this map which is a land cover degradation map in which from all these categories you only have three, whether it is degrading, stable or improving. And you can calculate the percentage and the area that is degrading or not. So as you can see for areas basing using this default data methodology wouldn't be very informative. Okay, so now let's move to the second indicator which is land productivity dynamics. After the live presentation we will work more on this. We will see other alternatives but let me just introduce that briefly what we mean by land productivity and land productivity trends, land productivity dynamics. There are many different subtle differences among all of these terms. Basically what we want to see are long-term changes in vegetation productivity. And this is, to measure this, we use, for example, the normalized different vegetation index which is a satellite derived index that is based on how plants different should be reflect different parts of the electromagnetic spectrum. So we have the near infrared and the red and depending on whether vegetation is healthy or stressed, they reflect these parts of the electromagnetic spectrum differently. Also, if you have more vegetation you will have a higher NDVI. And this is how it is, it's a normalized difference of these two variables. And NDVI is used as a proxy for many different vegetation properties. It's not necessarily exactly net primary productivity but it's a very good proxy and this has been studied and used for many years and it's very well established. And what is productivity important? Well, what is productivity? Productivity is related to the carbon that is fixed by plants. So it has a lot of implications for food security or climate change, et cetera. In generally, if you have more NDVI you have more productivity, you have more greatness, have more vegetation, more healthy and more abundant vegetation. So what we can do with satellite images is to go back in time and analyze time series of NDVI so that we can see what has been happening for example, for the last 20 years has there been an increase or a decrease? And for this we have, for example, this dataset from Moly's which is really good and available and easy to use in which we have data for every 16 days on NDVI and also on other vegetation indexes. It has a 250 meter spatial resolution and per year in one year since it is, we have one image every 16 days actually it's a composite we choose is the best available dataset in those 16 years. So since we have a map every 16 days for one year we get 23 composites. I'm sorry. Okay. So with this data, once we analyze it we will see that in Moly's depth we can derive different sub indicators. So we have 23 values per year and then we have many years. So we can analyze this time series of NDVI data in different ways. And for let's say, data in different ways. And for LVN and for this particular indicator which is lambda-optivity dynamics there are three sub indicators. One is trajectory, the other one is state and the other one is performance. Now we will see what these are about. And combining these three sub indicators you obtain a final indicator of lambda-optivity dynamics. So what is trajectory? As we were saying, for this is, for example, a year. Data of NDVI data for one year. So we have 23 data points, for example, for one pixel or one area. And we need to summarize these 23 values into one value which is usually the annual mean at least it is what is used in trends.org. For example, you can average this or you can obtain other metrics but let's say that we average this and we obtain an annual mean. So for each year we will have one value of NDVI. And trajectory, what trajectory is about is about fitting a linear regression to this time series. So here we have time and here we have the annual mean of NDVI and we fit a linear regression. We analyze its significance because we need to see if this increase is actually significant or not or if the decrease is significant or not in statistical terms. That's why we use the Mandel-Trend test. And if this trend is positive and is significant, then we say that there is improvement. If not, if it is negative, it's declining and if it is not significantly different from zero with the slope is not different from zero, it means that it is stable. And like this, we obtain this map of a trajectory. The second sub indicator is state. State is related to a comparison, is related now, is a comparison of two periods of time. So you need to define these two periods of time first, which is your baseline and which is your target period. For example, in this example, the baseline is from 2000 to 2012 and the target period is from 2013 to 2015. So then you classify your data in percentiles and you characterize whether in your target and your baseline period in which of these percentiles your mean NDVI falls and you make the difference. And if this difference is higher or equal to it's positive and higher to two, then you classify this pixel as improving. If it is lower than minus two, it is potential degradation and if it is between plus one and minus one, there is no change. So you obtain another map in which instead of with trajectory where we were looking at long term change, whether there was an increase or a decrease. With the state, we are comparing the last three days with the rest of the year because maybe there is a change that is only happening in the last, for example, three years. And with the long term term, with trajectory, you will not see it. So this is why this sub indicator is also useful to characterize number of changes. And finally, the third sub indicator is performance. We said that state was a temporal comparison. Performance is a spatial comparison. And the idea of performance is very interesting because it's trying to compare a specific area. So let's think of these as pixels to other pixels which have similar characteristics in terms of soil, in terms of land cover, and compare NDVI in these two different places. Why? Because we would consider that the maximum NDVI we see in all the places with similar categories is a proxy for the potential of that type of land, the potential for productivity. And even in your pixel, you have a very lower value than that, then your area is degrading and could give more in terms of productivity. So that is the basic idea of performance. So you need to first classify your area in terms of land cover and soil units. And now again, you see the distribution of NDVI, you find the 19th percentile because sometimes the maximum could be an outlier, so just in case, but let's think of this as the maximum productivity that that type of land can achieve and you compare it to your observed productivity in your pixel. So if this performance is lower than 0.5, your area is potentially degraded. And this is how you classify this. So how do we obtain this final land productivity dynamic map? Okay, and this is just the methodology that is implemented in terms of earth. There are other ways of also obtaining these land productivity dynamic maps that has five categories, as you can see. In the map, you cannot see them because in ISB, most of it is improving and also because recently trends.earth changed the color palette to make it for color blind people, but it's very difficult for not color blind people to see the differences. But okay, let's see how we obtain these five categories, which are, until now we were talking about improvement, stable and degradation. Three or two categories because with performance, you only obtain two categories. Combining these three sub-indicators, you can characterize land as improving, stable, declining, and these two new categories which are stable, but stressed and early signs of decline. And these are related to state and performance indicators. So if you have improvement, how do you read this table? So if you have improvement in trajectory for one pixel or one area, you have improvement in trajectory, you have improvement in state, and you have stable performance, you classify it as improvements in this free classes map or as improvements in this five classes map. But let's take a look at these two. If you have stable trajectory, that means there are no changes in the long-term trend. Now, when you fit a linear regression, the slope is not different from zero. Let me check the time. Yeah, okay. So we have stable trajectory. We have stable state. So in the last years, there is no chance with the respect to the baseline, but we have the tradition in terms of performance. What does this mean? This means that this pixel could perform better, could have higher productivity, for example, if it were managed differently, or because we know that other areas with the same land color and the same soil type have higher productivity. So you classify this as stable, but stressed. For early signs of decline, as you can see, it's related to state because you have a stable trajectory, a stable performance, but the gradation in state, meaning that the last years, the period that you define as target, you saw the gradation. So the decline is very recent. So it's early signs of decline. Okay. So this is how you obtain this map of land productivity dynamics using trends of earth methodology. And this could be used, for example, as part of the biophysical assessment in land use planning. And finally, the very challenging indicator, which is changes in soil organic carbon. We saw land color change, land growth dynamics, and changes in soil organic carbon. So basically what we want is to compare two periods, two moments in time regarding soil organic carbon, which is a key indicator because it's related to many functionalities of the soil and productivity. And of course, all these indicators are related. But this indicator is very important. The problem is that it is very challenging. All the people who work in stying soils will know what I'm meaning by this because soil organic carbon data is usually a legacy data, data, for example, that covers long periods of time. For example, in Argentina, our soil organic carbon map has data from the 70s until now. So this is a 50 year data set and you make only one map with all of that data. So how do you measure changes in soil organic carbon? It's very challenging. But so which is the approach that is recommended for this? A combined approach between soil organic carbon data and land cover changes. Because we know that the land cover, that for it in general, of course, for us, for everything there are exceptions, if you change from a forest to a grassland or to a cultivated land, there will be a decrease in soil organic carbon data. So combining soil organic carbon data and land cover change data, you can obtain the gradation map based on soil organic carbon. But as you can see, the reason it is also not very informative for Arias because you're using global data and it's also always very related to your land cover change map. So how do we combine in LDN framework these three indicators? So land cover productivity and soil organic carbon change. We have these three maps and we need to obtain one map that allow us to calculate these, which is the proportion of land that is degraded over total land area. This is the indicator for the SDG for the target for LDN target in SDG 50. So there is only one indicator, which is the 15.3.1 indicator. And this indicator is a number. It's actually not a map. It's just a number, which is the proportion of land that is degraded over total land area. And with this methodology, we are calculating this using a map, which is actually very useful for land use planning and also for management because we need to know where the gradation is happening. But of course, there are other ways to estimate this number without actually mapping the gradation. And for the combination of these three indicators, the one out all out principle is used. That means if for a given pixel, one indicator says there is the gradation, then in the final map, it will be the gradation. And you obtain the final map of this indicator. And you can calculate then, which is the percentage of area that is degraded, which for example, for high edge basin is seven square kilometers, which is only 0.6%. Using this methodology. If you use another methodology, you will obtain very different results. And this is where as experts, as planners, we need to use our expertise, our knowledge to actually find the most useful data and the one that makes more sense and is better for our study area. So now, the trends that gives you the possibility of not using global data for certain indicators. And it is very useful, very easy to use. And what do I mean by trends of health? Well, it was specifically designed as a decision support tool for reporting SDG 15.3.1 and it operates as a free plugin to compute GIS. And it also uses Google Earth Engine computing power and data. So we are kind of now looking at an example of using some tools that we have already seen in other webinars, QGIS and Google Earth Engine. So this just before we move to the demonstration, this trends.eth was developed by Conservation International with a project that was about enabling the use of satellite information, making it easier. You will see that you don't need to know much about GIS or Google Earth Engine to use it. It's just a few clicks and you get your maps. So Cesar, I will stop sharing if you are over there. Hello, good morning or good afternoon for you. Okay, I will share my screens wrong. Participant on the chat. Okay. Can you see my QGIS? Good. Okay, I will turn off my camera so I have more speed in my screen. So here is a QGIS. This is a project we did. We started with QGIS in one of the webinars and in the second webinar, we did this trends.eth, sorry, the Google Earth Engine. We're here, for example, we calculated this NDVI layer for IaaS for August 2020. So this is a project we have in QGIS. We are going to add now some layers from trends.eth. As Ingrid mentioned, okay, where does trends.eth comes from? Here, there is a web page. If you just look for trends.eth in your browser, you will be taken to the main trends.eth web page where you have a lot of information about all the things that trends.eth can do and how to use it. There are some tutorials and everything. So it's a very well documented. It's also, I don't know, some languages. So this is the web page. And how do you get this into QGIS? Well, this is a QGIS plugin. So basically, if you come to the, all the plugins for QGIS are, which is just here on top, you will be able to find trends.eth. So basically coming to plugins, you get this window. And here, if you search for trends.eth, you can find the plugin and simply install. I already have it installed. So I only can reinstall it or uninstall it, but you will have an install bottom and it's as simple as that. To use it, it's also not very complicated. There are here, once you install it, you will get in your QGIS this nice set of toolbox. For example, this first one, the setting tool, is where you can register. You have to register with a login just to have your email in the system because every time you send a task, as Ingrid mentioned, this over is connected to Google Earth Engine. So the computations are not done in your computer. They are done in the cloud. So after you send a task to produce a map, then they will send you an email telling you that the task is ready and you can proceed to the load. Here they have an indicator calculator, which is very, very simple. So far they have this land degradation indicator and also urban change and land consumption indicators and also some tools about carbon and biomass restoration that they are developing. We are going to go to the first one to the land degradation indicator. And here you can calculate these three subindicators that Ingrid just showed. You can calculate productivity. For example, the process is very simple. You can choose NDVI, which is in the land trend productivity. And then here you have all the options for this subindicator of land productivity that Ingrid mentioned. You have trajectory, which can be the NDVI trend. You can choose your time period. For example, you can go to 21 to 2019, if you like. Also here, there are an extra settings, which are the use of different methods to calculate trends. I think after my presentation, Ingrid will continue with this, but here is the bottom in TransAir where you can include some weather information to your NDVI trend calculations. There are different data sets that you can use for this. Then is the performance. You can also choose the years. You want to do the calculation. And for the state, which is the other subindicator, you can choose your initial period and comparison period. So this is all the setting you have for this indicator. And then you simply say, okay, where do you want to calculate it? I have here my IaaS shape file. So I can just include my IaaS shape file, click next, and then I can say, okay, this is land productivity for IaaS from 2001 to 2019. This is a task name. So I can identify the calculation. I want TransAir to do for me. And then I just click calculate. By doing this, I have submitted a productivity task to Google Earth Engine. So instead of going to Google Earth Engine, which is the software we saw in the last webinar where you can do really a lot of calculus and complex things, this is a way, a simple way to connect with menus, QGIS and Google Earth Engine. Okay, what did this task went? So I send the task. And here I have this sign of a cloud with an arrow. This is the new Google Earth Engine task icon. So if I click here, I will have here a window of all the tasks I sent in the last 14 days. If I put refresh, for example, here, I have this land productivity. IaaS 2001, which is already being sent to the server and is running in the system. So here it tells me the task is being run in Google Earth Engine. And I need to refresh the list and wait until it's finished. Okay, it's already finished. So it was a very quick process to do all these three indicators. Once it's finished, I can download the results. Let me add it here, land productivity. IaaS 01, 19, if I press save, it will download it from the Google Earth Engine repo. And now I have here my, let me turn off the NDI. Okay, I have the maps that Ingrid was talking and was calculating before. So the state, sub indicator, the productivity performance and the trajectory, sub indicators. So they already get load into your QCIS with the lesson and all the data that you need. Great. Here also, if we go back to this land degradation indicator, here I can also calculate land cover and the soil organic carbon sub indicator. Land cover has seen the show. It's also very straight forward. You just choose your target and your final year. And you can edit if you are using the ESSA, the space agency ESSA land cover. You can edit the definition. So you have a lot of different categories in this land cover, but you can say, okay, this category goes to other land. This is a wetland because at the end, we need to put everything in the seven classes that the UCCD is using for reporting. And here in the second tab, you also have this table that will determine which land cover changes are considered degradation or not in your land. So for example, if going from a crop land to a grassland can be improved or degradation or can be neutral. You can change this as you like. And for your specific case, you can define which of these changes is positive, negative or neutral. And then the use is very simple. You can give just the same area study area and running in the system is very quick. We have also the soil organic carbon indicator which basically operates in the same way. There are some parameters here in advance that you can manually introduce or you can use default global data for that. But anyway, here there is at the end one folder icon that says low data. Here basically you can import your own data. So if you have a better land cover from different time periods in a better quality, you can come to land cover here and you can add if you have it in roster or in a polygon, you can add your own land cover. You can edit the classes definition so you can put your own data here to do the calculations. The same goes for soil organic carbon and the same goes for productivity. If you use one another method like Ingrid will show afterwards to calculate trends in land productivity, you can just simply come and add your roster or your polygon with this category, one for declining, two for early signs. So you just add it here in the system and you can use your own data for the calculation which is totally recommended instead of using global data, especially for small areas like this. I already did all the maps here, Ingrid produced. So this will be the ESA land cover for 2001. And this is for 2018. And then I have the land cover degradation map. So once I have my three two indicators, I can totally come to this final part which says calculate final SDG. This already, if I already calculated indicator trends will automatically identified which are the different subindicators and already added here to the menu. So I only need to add the study area defined here. I can come to IAS data, put it a name, SDG for IAS and save it. And this will create different things. We'll create a few maps, but it also will create this table that Ingrid was mentioning. It is an Excel table that contains all of the information. So here it is the land productivity dynamic maps that you get with the global data mentioned before. This is a very course resolution global data. For the area, I think I have some of the original colors. I can load to the map, okay? So we can see it's mostly a productivity is increasing, stable, there are some places where there are early signs of decline and some places which is the client. I'm sure if you put a better land cover and a better soil map, you will get quite a different view. And here is the final indicator. As we mentioned before, you get these maps at the end of the analysis, but you also get, which is very interesting. Let me just use open the table, okay? There is an Excel table here. Okay. So you get this nice table with information about the final numbers of what is degrading, what is improving in the area. And also you get some analysis indicator by indicator. So for example, of the areas with different productivity trends. So areas that are, for example, in this transition, which are in early decline, you can see the land cover distribution of these areas. You have information about soil organic carbon changes, the changes in land cover. So you can see, for example, that grassland being converted to cropland in two square kilometers. And you can find here a lot of information of the different indicators, already very, very easy to statistics, all nice presently in an Excel file. And also for UNCCD reporting, you need some tables. And these tables are automatically produced in the here in the last tab. So you can upload them directly to the report. Basically, this is the information that you get by using TrendsEd. As you can see, it's a very, very simple tool, very intuitive. It also have some extra buttons over here. It has all the raw data. So you can also use it to download data that is stored in Google Earth Engine. They are here different, as you can see, indicators, precipitation data, soil moisture, soil organic carbon for different time periods. And you can simply get one of these and you specify your area and you can download the whole data set from here. So it's also good to obtain some data. This is, of course, global data, but for some variables, if you don't have local data, this is very, very helpful. And at large scale, it's also good. Also here, you have another bottom where you can plot data. So here you can run different plots. There are NDVI. You can choose, we can try to do one. This will go very quickly. So you can choose one area and try to do a plot about the trend. So you can see how the trend is going. Perhaps it takes a while with my internet connection. Okay, let's see. Refresh list. Okay, it already finished. So I can download and I get here for example, this is an average NDVI time series analysis for the whole IAS area. So this is why the map is green because as you can see for the whole IAS basin, there is a positive trend and it's going up. Okay, so there are different tabs and button you can try. Ingrid, I think we show now different methods that you can include in your calculation. This will totally yield different maps. So it's very advisable that you explore a little bit more on what are the results you get using different parameters. As I mentioned before here, instead of just the NDVI, you can use the pixel restrain or the water use efficiency and you have here some climatic data sets that you can put into the calculation. IAS is quite a small basin. So for some of these huge global data that you get from this weather product, it may get very big pixels but you can try and quickly see how that will change your map or not. Then if you see there is a linkage, you can try using your own data, building a more high resolution approach. So I think this concludes my quick demo on Trends Earth plugin. You have the web page, you can Google it and also in your QCIS you can download the plugin and use it very quickly with the data is already available in Trends Earth. But remember, if you have your own data of better quality, use that and you will get always a better result, especially in very small areas. Okay, thanks a lot. And if there are any questions. There was a question Mustafa asked about organic carbon data. Mustafa, I don't know if you would like to, if it is enough with what we answered in the chat or you would like to ask something else? I see it. Yeah, this is global data. You can for example get this. Actually, I wrote my question. I think you may respond it to my question. You can ask your question verbally as well. So, okay. Organic carbon maps. We collect soil samples and we used other parameters when preparing such a map. We base our analysis on organic carbon content in the soil. Content in the soil. Are there different parameters to calculate it? Are there different methodologies to calculate the soil organic content of the organic carbon content of the soil? Sorry. Could you hear me? Yeah, well, actually the indicator in this context is changes in soil organic carbon. So it's even more difficult. As I said, this is a very different indicator to calculate because soil organic carbon data, it's usually legacy data. So, when you build your, if you have data on soil organic carbon, that of course you first, we will talk about this also in our next webinar in which we will talk about digital soil sampling and how you can predict soil organic carbon for other places in which you didn't measure it and which other parameters you consider, you usually consider. Nowadays, with all the machine learning algorithms that allow us to use many variables, we can make these models to predict soil organic carbon, including many different parameters from the texture of the soil, the productivity, the NDVI. And as much variables as you have available, you can use them in your model, of course, within that they make some sense for predicting soil organic carbon. But in this context that CESA just said in Trends Earth, and as I put in the chat, you use this soil grids map of soil organic carbon. But for example, you could use the global soil organic carbon map of FAO, which is available and you can download it and also use it as input or your own soil organic carbon. And you use this data of soil organic carbon in this context as a reference because you usually do not have two periods of soil organic carbon data for two times. So the changes in time, you evaluate with land cover change data. And you use your soil organic carbon data as a reference to see, to have a reference on the value, the magnitude of these changes in terms of soil organic carbon data. But it will not make much of a difference. Unfortunately, if you use different soil organic carbon data for this, land change data has more impact in this indicator than the soil organic carbon data itself. There is a lot of room for improvement regarding this indicator. Yes. Well, if I may, soil organic carbon is to be measured, is to be measured, not anticipated. It has to be measured, not anticipated. And as we measure the SOC, we have to look at the depth of the soil, the climate conditions, the vegetation type of vegetation, we have to look at them all. And geological, geomorphological structure of the soil is to be looked at as well. So a program cannot just be based on the satellite images. We cannot just calculate a chemical thing basing ourselves on a satellite image. And secondly, if you have a previous SOC map, then we don't have to measure the carbon once again. Well, if the map is relatively recent, then we don't have to measure it once again. If we have measurements at every 250 meters, as it's facing, for example, if we have a recent organic carbon map, then we can just use it. But I know this is not enough. And the live soil, the viable soil layer should also be included as a parameter. So lots of parameters are to be used, have to be used. We have FAO carbon map. I know I was engaged in the production of that map. I very well remember those periods. And we got included in that project with 8,000 samples from Turkey, but 78 million hectares of land, you cannot just cover all those lands with 8,000 samples from Turkey. I know that. So 1,800,000 is the scale of the map, I believe. So what I'm trying to say is, we can't just base ourself on a satellite image when trying to calculate the soil organic carbon. This is why I wanted to pose my questions to the speakers. Well, the presentations were very well, by the way. Thank you. But I was curious whether you had a different method in mind in order to measure the carbon. I just wondered whether you had a brand new methodology in mind to calculate the carbon. That's it. Thank you. Okay, thank you, Mustafa. No, of course, yeah, the soil organic carbon mixed to me measured in the soil. And as you just said, the global soil organic carbon map of FAO is based on soil samples. And as so is the soil grid soil organic carbon map. But if you need to map it in a continuous area, you need to somehow interpolate it. You need to interpolate these data points. Actually, if you measure it in one point, in one specific area, which is where you took the sample, the soil sample, soil organic carbon will vary a lot. And if you measure it again in 100 meters away, it will have a very different value. And so you need to consider, it's impossible to have enough to measure the whole grant a wall-to-wall approach. It would be great to do that, but it's impractical and impossible at the moment. Maybe someone, not me, will come up with a better methodology, quicker, cheaper, and easier to measure soil organic carbon. And this is why it is so important to when we make maps of soil organic carbon, which can use satellite information for making better predictions or not, but you don't only use satellite information to improve the predictions. You also include other types of information because, for example, maybe you can use soil or geological information that was obtained not from satellite data information. The idea of using all these other variables is to make better predictions because you can interpolate your data only by using special autocorrelation methods, like creating, for example, normal simple creating, and you just interpolate your data and predict your data using this information, the special autocorrelation information, or you can use other methods that, for example, you can consider. For example, if there is another land cover, if there is another soil type, if you have all this information, you expect that you will make better predictions using this ancillary information. But of course, the predictions are based on soil carbon data, as you said, that you need to measure from the soil. And what I want to say is that this is a great issue to discuss, and it also highlights the importance of mapping uncertainties when we map soil organic carbon. I'm considering this because it gives us an idea of how far we might be from reality. And sometimes this is not considered, and all these new techniques allow us to measure uncertainty also, and accompany maps of soil organic carbon with uncertainties. Cesar, you were trying to say something? Yes. I was also like to speak about an answer to the question. I was thinking because the idea of this indicator is that, is that you can be able to measure the change in time of soil organic carbon, to know if you are losing or you are gaining soil organic carbon. So the idea of the indicator is very simple, which is hard is to make it to know what is really happening. So this is the most easy approach that was proposed by UNCCD, say, OK, we can start with this map and we can start. You actually are looking at changes in land cover, not soil organic carbon. We put there the SHISOAP map of FAO for many projects and you get the same result because this is only map of one period, and then you see the change in land cover. But I guess for some countries or for some places where you monitor soil organic carbon or you have all measurements and new measurements, the best approach will be to make two different maps. One map of a previous year with measurements, I don't know, from all the samples you can get until 2000, and one new map with all the new samples. And then you can put here in this tool the two maps that you prepare with soil samples of different times. And in that way, you can try to estimate, make a better estimate of the change in soil organic carbon. But the only way of really accurately do it is that if you have a monitoring system where you measure soil organic carbon content every five or 10 years in the land. For a country, it's sometimes very hard to do because it's costly, and as we said, we need a lot of samples. But for small basins, you maybe can. Yep, go ahead, Mustafa. Yeah, thank you very much for all the nice information that you have provided. But the thing is, the phosphate or potassium in the soil, can we also calculate them this way? No, the carbon is going to be used in organic matter calculation in a way. So especially when we are making recommendations about fertilizer application, we are using organic carbon as well because we need a parameter in soil which will not be changing in the soil. The nitrogen will be changing very much in the soil under different temperature circumstances, but you have to find something which will not be variable a lot. This is why we are basing ourselves on organic carbon when we are recommending fertilizers to the people. So the thing is, if you can just benefit from the past data and also satellite image, and if you can end in a map, then this is so nice. We can use it for the productivity indicator as well and also when identifying the nitrogen amount in the soil, we can use a similar tune, I believe. So what I'm trying to say is, what is the resolution of the map that we are going to get from this tool? What is the accuracy rate using this tool? What is the resolution scale? What is the rate of resolution of the map that we are going to draw from this tool? I'm not just trying to reject or do anything, I'm just interested. Can I intervene? Emra is saying, Hakka is saying, can I intervene? Okay, please. Can I hear me? Can you hear me? Says Emra, yes. Mustafa, just clarify if I may. UNCCD prepared videos about each and every indicators, indicators presented today and about the resolutions of the maps. There are these videos of the UNCCD. But the thing is, these global data in trans-Earth, the things about soil organic carbon and the FAO map and the soil grid, these are all global data. These are all global data. And as Cesar said, they generally suggest a reduction and the increase in the carbon. This is the idea behind this indicator. And as a country, which is a party to many conventions, we are obliged to do reporting about reduction and increase in the soil organic carbon. And the ultimate idea of these indicators was to identify the baseline of the carbon of the countries at the beginning. This is why from now on, when you do reporting as a government, as a country, you are going to see the differences in the trends when it comes to organic carbon. But first of all, you have to see the baseline, right? This is why a FAO has this global SOC map. It is acting as a baseline. And it was prepared as a baseline. And in soil grid, the soil grid map, at the scale there, one... I can understand the scale, I'm sorry. But I know about the scales of those maps. I was involved in the sampling process of those projects that you are mentioning. But in general, those maps, use of those maps in a local, in a small area as Ayash, may not be so possible, I'm afraid. But those data, those global data, and maps can accompany the regional data, local data from Ayash. This is how you can make use of the global data. Global data, can accompany the local data, both soil grid or FAO's SOC map. These are global data, and they cannot be used as a standalone data in Ayash basin. Even if they are used, they will not be giving detailed data to you, unfortunately. That is why global data, how to be enriched with regional and local data. This is what we're trying to do in this project. We are trying to draw some local regional data from the labs laboratories as well. Please do not consider that global data will be used as a standalone data in this project. They will be just acting as the baseline, global data, and they'll be suggesting certain trends in increase and reduction in the soil organic carbon in certain regions. But they'll not be helping us with an actual, with the very planning process. No, local regional data will be required. Data from the region, from the field, will be so much important. That is it. I think Cesar would like to say something. Yes. Yeah, no, just a quick comment before I leave England with her last presentation. But I, no, I wanted to comment on that thing of the scale. I mean, this is just a tool, and most of this has Haki say was thought for country reporting. So the idea of this is something that can be implemented by every country at the country level. So it's very high scale. And that's what I constantly try to mention that if you are in a small watershed, you need to put different data, your own data, and do it differently. You can use these tools to help you with the technical part of mixing the satellite layers or the maps that you produce, but you need to use your own local data for such a small area. But, okay, I think England has another presentation and we already made it right. Sorry. I should continue. I will not say anything. Actually, now we will see, for example, that an alternative way to measure productivity changes and using another methodology. And as we were just saying, also trends of earth gives you the possibility to upload your own data at the scale that you want to work with. So the idea is to present these tools, for example, but you can not necessarily use default global data. You can use trends.earth and use your own data, especially for land cover and soil organic carbon. So I will continue so that we finish I think we should be finishing by now. But anyway, I think it's very good to have these discussions. It's actually the best part. Oh, I should go. Sorry for this. It's good to remember all we have seen. I put the slide we were supposed to start with, but I don't know why. Okay, now we are arriving. Land productivity trends. Why are we focusing on this indicator? Well, because it's the one that in general, in many countries is the one that is the most informative one. And since we saw in soil organic carbon and land cover changes are very associated for reporting or with this methodology, if you make a change of, for example, the way you're managing land, but you do not change land cover, that will not be reflected in these indicators. Of course, for LDN, these are just some indicators proposed for monitoring and reporting at national scale. And we need to learn from this and change it. And also many other indicators need to be included in our analysis, other types of indicators at national level, at regional level, regarding other socioeconomic indicators, strength reduction indicators. So this is not just the only thing that has to be reported. And the idea of the indicators is to help us, not just reporting, is to help us address land degradation and make some changes. So we should always remember that. So we, for example, we were talking about NDVI. Is NDVI the only vegetation index that we can use? No, there are other vegetation indexes, for example, such as the Enhase Vegetation Index and the Soil Adjusted Vegetation Index that could be more informative for a particular region or area. For example, the Soil Adjusted, the SAVI, the Soil Adjusted Vegetation Index, it's better, it's more sensitive when the vegetation cover is low. And on the contrary, EV is better in areas with dense vegetation. And these are available, these indexes are available and we can use them. So this is another way of also improving. Sorry, I think I'm out of the microphone. Thank you. Okay, so that is one thing I wanted to talk about. Another thing is how you, we talked about the 23 values per year and how we integrate this, how we summarize this annual information. Well, we are in Gens.Earth, you calculate the annual mean, but you can calculate other metrics, which could be very informative to and address different things. For example, with NDVI data or EV data or SAVI data, you can calculate, for example, different phenological metrics, such as the length of the growing season, the rate of growth, the rate of innocence, the time where it starts, maybe there are changes in the, long-term changes of when the growing season is starting, maybe it is delaying or starting before. You can also characterize the seasonality and see if there are changes in seasonality over the years. For example, instead of calculating the annual mean, you calculate the coefficient of variation. And there is also another index that is called SP, that is the annual ecosystem services productivity index, which is the, it considers the annual mean, the NDVI annual mean and the coefficient of variation. It considers both things. And here, for example, you can see, these are different years. And here for the same NDVI data, if you calculate the annual mean, which is in red for every year, you get this time series of annual means of NDVI. So there is a 17 year period here. And here, around 2011, there was a change from grasslands, shrublands to cultivated land. And the mean NDVI will not change that much between these two types of land cover, but the SP will capture more, will be more sensitive to this change in land cover of land use. And because it not only considers the annual mean, but it also considers the coefficient of variation within each year, the seasonality. Because cultivated areas may have the same annual mean, but they will have a higher seasonality than grasslands. So for example, instead of using time series of annual means, you can use time series of SP, which would give you more information. And we did that. We did that for Argentina. We did that for Upper Zacharia basin. For example, here is a paper in which we compare different methodologies, different analytical approaches, and how very different results you get, depending even with the same data. You can use different data, but you also with the same data, if you use very, if you use different analytical approaches, for example, this column is using annual means, and this column is using SP. And the rows correspond to different ways of characterizing the trends, different analytical approaches. Everything is in the paper. We can discuss it, I will not bore you now about this, but the bottom line here is that even with the same data, with different analytical approaches, you will get very different results. And how important it is to... This, when we were discussing with SOC data, it's valid for every indicator. These are, for example, these six maps for a, for Irish basin. And as you can see, you get very, very different results. These are the three different methods, the long-term trend we call the SWATI and the S SWATI. These are just different ways of analyzing the data, which are all valid, but which is the best when this is the key question. For this, we need to validate maps, we need to ask experts. And for example, we also calculated a consensus map from these six different maps. And as you can see, a little bit more of area is degrading. It's almost not 2%, it's 1.7%, but it is a bit more than we have seen with the other methodology. And also how you can characterize the magnitude of these changes. Now, we were talking about magnitude. It's not the same to say, oh, it's degrading, but if it is a very strong negative trend or if it is a light negative trend. So this is also an issue that is important to quantify the magnitude of the changes. This is what Cesar was talking about when he was showing trends.f that also, this is a big discussion in the academics of whether you are interested in distinguishing if the negative trends or positive trends are related to human-induced changes or climate variability. Because primary productivity is affected by many things, water, light, temperature, and you need to interpret variability using historical precipitation information as a context. And however, for LDN, it's important to measure the relation whether it is human-induced or climate-induced. This is important maybe to analyze, of course, but it's not necessary to discriminate. And also, the climate correction methods that I used to include historical precipitation information in the analysis will depend on this information. And if you have available information, the historical available information is not good, then you will have very bad results too. I think we do not have time to go in detail on which of these methods. There is the residual trend analysis. Here is a paper, the rain use efficiency, trends of rain use efficiency, or you can calculate trends on water use efficiency. And here are the available data sets in terms of earth for historical precipitation and evapotranspiration. And we saw how to obtain these land productivity dynamics maps. Remember when we were combining these three sub-indicators and we obtained these five class final maps. Well, there are other ways also to obtain a land productivity dynamics map. This is, for example, the map for saccharia-based, for upper saccharia-based using the JRC simplified methodology. And you can read more on this, but in this methodology, another approach is taken, but you also obtain these file categories, including early signs of decline and stable path stress. And also you can be creative and think of a methodology that better represents what is happening. And this is also something that I have to say always that in line with what Mustafa was saying, we need field data to validate and to have information that is actually represented what is going on in the ground. We cannot rely simply on satellite data, not only for solar-organic carbon data, but also for productivity. And we are making a very simple association between positive trends are improving conditions. Is that always true? No, not necessary. In this area, for example, I really like this picture from Hans Peter Lieniger. He is showing here an invasion of a woody species that of course you will see an increase in NDVI, but this is not necessarily an improvement because if an invasive woody species is invading everything and cattle cannot use it and biodiversity is decreasing then, even though you have a positive trend, this is not an improvement in terms of land degradation. So we need to go to the field and we need to ask experts also about their knowledge on the area. So to finalize this webinar, some final remarks, let's not forget, we are talking about land use planning and integrated land use planning is a great way to balance environmental, economic and social priorities which all need to be taken into account also in LDN. And what we need to focus is using the best information available, whether it is global or whether it is regional or whether it is national, let's use the best information that is available on land degradation status, but also land potential, social, economical data and general considerations and optimize the spatial mix of possible integrations. And this is why E-loop is key to achieve LDN. And it is always important to emphasize the importance of multistakeholder participation for making these processes successful and useful. So that is a, yeah, thank you very much. And I don't know if we have, I think we are already past 10 minutes, but we started also 10 minutes late. So I think we are okay with the time. Thank you very much. If there are any questions or comments, now it's the time. Is there someone with a question, Mustafa? No question for now. Thank you very much to your colleagues for the presentations. As for the soil mapping, we are waiting for a more efficient webinar on soil mapping. Once we get that webinar, we will have more questions to ask. I guess the mic of the head of the department is still glitching. I guess there is no question. I'm trying to reach the head of the department through phone. I'll get back soon, very soon. Please wait, hold on. Thank you very much, Ingrid and Cesar for these presentations. Thank you. Papa, you're making me be more scared and scared of the next webinar. Okay. Thank you, everyone. Nice to be here. We have so much to talk about. Very special on the microphone. I guess the mic of the head is still glitching. I guess we can now end the session and the webinar. Thank you for participation. Hope to see you in the next webinar. I'd like to thank Koyal and Denise for the interpretation services. Thank you very much. Stay safe. Bye-bye.