 Good afternoon, good evening, depending on where you are connecting from. It is my pleasure to welcome you to this new webinar on Glozolan soil spectroscopy. My name is Julia Stanko. I'm sorry that my camera is off, but I experienced some issue with my camera settings just now, so I switched it off. I work for the FAO Global Soil Partnership and our webinar today takes us to Denmark, where our renowned speakers will introduce us to soil spectroscopy from an academic and private sector perspective. We are very lucky as they will also show some practical examples of research projects and laboratory demonstrations. Before starting, I would kindly remind you that the session is organized in a webinar format in which participants cannot activate their audio and camera. However, participants are encouraged to post their question in the Q&A box, which will be moderated by my colleagues. We will choose a few questions to be answered live and the rest will be answered via chat. In addition, a chat box is available and can be used for interaction between participants. For any technical issue, please write early to me on the chat and I will be happy to help. Before digging into soil spectroscopy with our renowned speakers, I will provide a bit of background on the Global Soil Partnership and Glozolan, the Global Soil Laboratory Network. So the Global Soil Partnership towards healthy soil and sustainable soil management for all. What is the Global Soil Partnership? The Global Soil Partnership is a single and strong voice on soil issue. It is a mechanism and a partnership created to improve collaboration between all stakeholders who dedicate their work to soil. It is a unique framework for exchange of experiences and discussion among all actors and is composed of seven regional soil partnerships, over 500 partner institutions worldwide, seven technical networks and 161 national focal points from the 197 FAO members. Public and private donors are supporting GSP action on the ground. Here you can see our ITPS, the Intergovernmental Technical Panel on Soil, 27 soil experts providing scientific and technical advice to the GSP. The Plenary Assembly, the GSP main decision-making body that held its annual event with FAO members and GSP partners to review and prioritize soil action. And then you can see in my slide also the areas of work of the Global Soil Partnership. Glozolan, the Global Soil Laboratory Network is one of the seven technical networks of the GSP established in 2017 to build and strengthen the capacity of laboratories worldwide in soil analysis and to respond to the need for harmonizing soil analytical data. 2017, here you have a short timeline. 2017, Glozolan started to work on wet chemistry and focus mainly on training, standard operating procedures, and the execution of interlaboratory comparisons. In 2020, there was the launch of the Glozolan Initiative on Soil Spectroscopy. In 2020, the related launch of the International Network of Fertilizer Analysis. And then here we are, 21-22, consolidation of the capacity development program and activities on spectroscopy through training, manual, assessment, webinars, and video courses. So here we are now. One second. Okay. The Glozolan Soil Spectroscopy webinars. We had many also in 2021. This is the new cycle of 2022. I will now give the floor to my colleague who will be moderating this session and introduce, we'll have the pleasure to introduce our today's speaker. E over to you. E, you are muted. Oh, sorry. Thank you, Nicola. And thank you, Julia. Thank you very much for this brief presentation to introduce GSP's work and what we are focusing on. Hello, everyone. Welcome to the second Glozolan Spec webinar of this year. My name is Yipen from Global Soil Partnership, FAO. I'm coordinating activities related to Soil Spectroscopy, soil data and information in general. As probably you know from our previous webinar that one of the main objectives of Glozolan Spec is to help countries build their capacities on the development of soil spectroscopy in order to facilitate soil monitoring and sustainable soil management. This is also why last year we decided to run this webinar and invite advanced institutions to share their knowledge and experiences on the topic of soil spectroscopy and soil mapping. Today we have invited our colleagues from Arcus University, Denmark, to share their stories during the last 15 years, how they developed their soil spectral lab, National Soil Spectral Library, and how they explored the capability of this technology in soil application. Today is also a bit personal for me because I did my PhD in this group 10 years ago and closely worked with Dr. Maria Canada since 2011. In the meantime, today's webinar is also slightly different compared to before because we will have two speakers. Another speaker is Dr. Nikola Boek from one Danish spectrometer manufacturer called FORCE. Since 2013, it was the ending period of my PhD study, actually. Arcus University started to work with FORCE for further development of soil spectroscopy at the national scale. So we worked together to build the second generation of the Danish soil spectral library, developed a national model for different soil properties. Until now, Arcus University and FORCE are still closely working together on the development of soil spectroscopy. Again, this experience is also a bit personal for me because I was in the part of this journey back on that time. So I really appreciated this experience as well. So I think it would be really nice to share this story to the world, how Research Institute can work with a spectrometer manufacturer and develop their capacities for different soil spectroscopy applications. Without further ado, I will pass the floor to today's first speaker, Dr. Maria Canada. In the meantime, we would like to introduce Dr. Cecilia Hammerson. She will be in a panelist today to help with you and your questions. So please feel free to raise your questions in the Q&A box. Maria, floor is yours, please. Thank you very much for this very nice introduction and hello, everyone. Let me just put up my presentation so you can be able now to see the first slide. Yes, go ahead. Yes. So first of all, I would like to say thank you very much for the invitation. I feel really privileged to be a part of the training workshops that you're organizing. I think it's a really, really nice initiative. It's also nice to see you here in slightly different constellation after some years. So yeah, I'm here to share our experience with soil spectroscopy and application of spectroscopy to different activities we have in our department. And as she already mentioned, I have Cecilia Hammerson, my good colleague with me, who will be helping answering the questions during the presentation. And she has been also involved in a lot of spectroscopy activities we have. So during the presentation, I will first start to tell you a little bit about the department where we're working and why we're interested in soil spectroscopy. And I'll also show you the instrumentation that we have available and how we are using the instrumentation. So the dataset, the spectral library that we have developed. And of course, the aims of the library are to use them, the data sets. So I will show you some examples of different applications. And I will finish telling you a little bit about our teaching activities. The Department of Agroecology is actually located in different areas of Denmark. So we have research facilities in Follum, which you can see on the picture here. And also in Flakebia. And we have teaching activities mainly in August. And we have also research stations in other places. So we are distributed in different areas of Denmark. The main activities, of course, are research. We are part of university. And we also have activities related to policy support. And this is where we get certain questions from the ministry that we have to answer. And sometimes we need to conduct research, further research in order to be able to answer them. So this is also quite a big part of our work. We also have education programs for bachelor students and masters within agrobiology. And we are also partners in the international Erasmus Mundus master program within soils and global threats. And finally, I mean, we do a lot of collaboration. And this is, of course, not only in Denmark with other institutions, companies, as you will hear in a moment about other universities, but also a lot of international collaboration. So when it comes to the staff, agroecology is 270 employees. We're around 60, I think, are PhD students. And the group, which I'm affiliated with, we are 40 and total and eight senior researchers. Within the department, we have several different sections that are working with, with different topics related to agroecology. So I've just listed them here, but I will not talk more about this. I will instead focus about the soil physics and hydropedology group, which in short we call soil. This is the group that I'm affiliated with. And within this group, we also have a lot of different activities. We do research within arctic soils, digital soil mapping, soil quality, soil mechanical behavior, also sustainable soil management and water and contaminant transport. And finally, we also work with spectroscopy. And I'm responsible for the activities within this area. And here we work with NIR spectroscopy, which is the main topic of this presentation. But we also work with mid-infrared range and laser induced breakdown spectroscopy. So the group that is working with spectroscopy is quite dynamic. At this point, we have some master students, PhDs and postdocs. And since it is a lot of students that are involved, then of course, there is quite a dynamic because students come and go sometimes stay. So we do have, like I said, postdocs as well. So why, why at all are we looking into spectroscopy? I think it's a similar reason why everybody else does. And this is because we simply need more and more soil data. And this is on different scales from local field scale to original and national. And it's difficult to catch up with this demand when we use only conventional ways of analyzing soil. So we go, of course, after some other alternatives, in this case, sensing techniques. And in the department, we use different sensors, not only spectroscopy. I'll be talking about NIR only. But we also want to use this technique to actually be able to estimate some properties, maybe a little bit better. And this is mainly for properties that otherwise in the lab would be difficult to be estimated repeatedly with the same precision. So rather complicated properties. And the last two points I put here are rather related to proximal soil sensing in the field, or remote sensing, where we are interested in gathering more data on a bigger scale, or maybe in the areas where we cannot access. So in the department, we have different instrumentation available. We have spectrometers that are used in the laboratory and in the field. In the lab we are using mainly two. We have DS2500. This is the process instrument. I will not talk too much about it because Dr. Nikola Bork will introduce you more to this instrumentation and we'll show it as well. But this is, in short, it's an NIR, visible NIR, sensor gathering spectral range from 400 to 2000 nanometers, 2500 nanometers. And we are using this special sample cup for measurement. And we can do measurements in several locations. So the sample is actually rotating while we are taking the measurement so we can get a nice representation of the soil. The next instrument is a lab spec. It's quite common actually among soil spectroscopists. It's an ASD instrument that we don't use as frequently and we usually use it when we need to do measurements on intact soils. Because we have this device called Probe, contact probe. For field instrumentation, we have various shank and profiler system. So it's the same company that has manufactured them. The first one here mounted by a tractor or pulled by a tractor. It's on the special toolbox. And it's actually quite special model where we have not only NIR spectrometer where we do the measurements through the shank, which is lowered in the soil here. We have a small stuffer window at the bottom through which we illuminate the soil. We also have coutel electrodes. So that's an additional sensor that we can use here. So we get readings for electrical conductivity. And we also have a temperature sensor. So in the same time while driving, we can record three types of sensing measurements together with 50s coordinates. And the profiler system, as the name indicates, is a system that is doing measurements on the profile scale. So this is where we are inserting the soil and doing measurements while inserting. Here you have the small stuffer window. We have not been using this system that much. We have a bit of problems with sandy soils and some of the soils where we have a lot of stones. So it's difficult to actually insert the rod in the soil. So this is the instrument making. Of course, for all those instruments, we have developed our protocols on how to prepare the soils, how to set up the instruments, how to extract the samples after the measurements, information on standards that we're using. So this has been years of work of different people involved in this. So let's go to some examples. Here, I would like to talk about how we established a spectral library of sandy soils. So at the very beginning, we have selected some soils from our soil archive. We are very fortunate because we have thousands of soil samples in our archive. And they also have wet chemistry available. So we have selected some representative soils and scanned them with ASD instrument and sort of established this first database for analysis. And we have conducted several studies using this data. So I will show you one a little bit later. And in the next phase, we have actually re-scanned this library with FOSSUS instrument, DS-2500. And this was in relation to our collaboration project, which was dedicated to the application of soil spectroscopy for soil quality assessment. And in this project, we had August University where we had already some experience with spectral libraries. We had soils, we had reference data, and FOSS had the instrumentation. And we also had a third partner, which was a Danish consultancy company for farmers, which was supposed to disseminate the results of the project so that we could also inform, for example, end users, farmers, and advisors about this technique. And so we were trying to be as visible as possible and attending conferences where we knew where farmers are also coming. So we have prepared, we cannot see it, and it's actually in Danish. This was a poster that was summarizing for the farmer's idea about soil spectroscopy and how to use it. And we also made demonstrations using the sensors showing how quickly we can make estimations of basic soil properties. So the spectral library is quite extensive, if you think about Denmark, which is a very small country. So we have scanned several datasets available in our soil archive. I put them up here that is colored according to soil types in Denmark. So you can see the distribution of the points and how well we are actually representing the entire country. So yeah, we have soil profile investigation. These were samples from 1980s. And these are the samples on a several kilometer grid. These are small points you can see across the country. And they're also at different depths. So it's actually approximately four depths per point. We have data from soil classification database. It's highly dense points you can see here. Where we have points at two different depths, topsoils and subsoils. We have also a lowland database. It's actually not on the map, apologies. And several field investigations, the bigger points you can see here, those these are fields where we have intensively sampled. And we also have samples from other countries from Greenland, for example, we will keep on actually adding more from Greenland because we have quite many activities there. And also from other countries around the world. So this is our sort of basis. And we keep on actually enlarging the spectral library with every new project that we are running. So let's go through some examples because I guess that this is most interesting. This is this is one of the first examples of the application of the first database that we established on LabSpec. And this is where we use the square grid data and on the topsoils on the spectrum information. So the idea was that we wanted to see if spectra only can actually tell us something about soils in Denmark. And we perform principal component analysis of the spectrum information. And we got PC scores from the first three pieces that we have mapped. So we have a map of PC one, PC two and PC three. And already here you can see that there's a very nice there are some trends behind behind those colors. And if you correspond these maps to the loading from the principal component analysis, you could see for example that the first PC has quite high loadings within the region that is correlated with organic matter. And this is also what we know that we usually have more organic matter in this area of the country. Oppositely, for example, PC two has let me see PC two is the red line here. It has responses from OH bonds and some iron oxides. So mainly from clay mineralogy. And this is also what we know about the initials actually, or the sample PC maps could could tell us a lot. And the last PC three was a little bit of both. It also didn't explain as much variability. So I will not discuss it in details. In the next step, we have tested these three maps and got this sort of very simplified cell classification map of Denmark. And you can see also here, there is a plot with mean spectrum from each of those clusters. And these are consumer moved spectra. So it's not a row reflectance. And just as an example, again, the blue cluster here or dark blue brother had responses from iron oxides and the highest responses in areas where we know we have information from clay mineralogy. And if we take for example cluster four, the green one here, it has the highest responses from organic matter. And if I put again this tiny map of any so you cannot see similarities, which correspond very well to what we already know about Danish shows. You can say, of course, that this very simple classification map from NIR is much simpler. But this is also because we used on me approximately 600 points to come up with this map, whereas this map is based on, I think about 30,000 points. So it's much fine resolution. So this was just an example of more qualitative analysis using a spectrum. And from now on, I will show you some examples with quantitative modeling results. This first one is where we used NIR for the estimation of basic soil properties. And the soil properties that we know that are spectrally active in the near infrared range. So it was a study where we had quite many different fields across Denmark, or actually one field from Greenland you can see here, the extra triangle, what extra classes were represented. And so we have modeled clay content, organic carbon content. And the plus that you'll be seeing you can you can see that in the x-axis, there's always reference values and y-axis would represent the predicted values. So again, the first model for clay content, I will not talk about r-squares or errors, they are actually on the plots, but it's I would rather that you focus on the fit and how the samples behave along one to one line. And so organic carbon model, both good fits, there is a bit of issue here in higher range of clay content, but the overall model is good. And we added also here a model for dexter N, it's a ratio of clay over organic carbon, which is a very good indicator for soil fertility. So it can tell what conditions, what structure or condition the soil is depending on the number. So if N is below 10, we have non-complex organic carbon present, and it's rather good structure and good tilt conditions. And oppositely, if the N values above 10, we have non-complex clay, and it represents soils that that are of degraded soil structure. So you can see actually that the model that is representing the clay-to-ocean ratio is even better than the other two models, which was very interesting for us to find out. And we hypothesized that that's most likely because in this model, we are using information from both soil organic carbon and clay. The next example is from a project where we wanted to push spectroscopy a little bit further, and instead of looking into properties that are spectrally active, we wanted to estimate rather soil functional properties, which are dependent on spectrally active properties like soil organic carbon and clay contour or mineralogy. So in this case, we wanted to estimate different soil properties that are important for risk assessment of contaminants transport. So all the properties that are telling us something about soil filtering function. And I have listed for you some of them that we have estimated. So these were structure properties like microporosity, city metrics, service properties like specific service area, water repellency, transport, and binding as well, including sorption coefficient. And I forgot to tell you from the beginning that every time I talk about a study, I put actually an asterisk with a number. And at the end of the presentation, we will have a list of all the papers where we have actually published these results. So now I'll show you some examples from this project for the structure properties. So I mean structure is defined as the spatial arrangement of pores and particles and networks in the soil. And it's controlled by soil properties for climate, soil crop management as well. It's a very important property to know. And in this study, we have used an X-ray computed tomography, CT scanning that you might know from the hospital to derive the soil pore network status. And you can see here the picture of the soil column. For example, universal soil that is the sandy soil, very few pores not well connected and volvia soil, which is might well better structured with many pores well connected. So we have derived two parameters here, city metrics, which is a density of soil without depth pores and without the stones and micro porosity, which consists of pores bigger than 1.2 millimeter in diameter. And here are the results. At the top, we have modeling of macro porosity. So again, measured versus predicted and the corresponding regression coefficient. And at the bottom, you have a model for city metrics and the corresponding regression coefficient. So you can see again without any numbers that macro porosity was not very well predicted. It was actually not a reliable model. Excuse me. However, if we compare the results from this model with Peter transfer functions that are usually used to derive this property, spectroscopy still performed actually better. For city metrics, you can see that we actually were able to establish a very good correlation. The next action property is the specific surface area. And that is also very important. And it governs a lot of other processes in the soil are important for water retention and movement also for contaminants, dynamics and nutrients as well. And so here we use the big big set with a lot of samples actually from different continents. There you can see again, we have a very nice distribution if it comes to the texture. And we have divided our data set on calibration and validation and the top model is actually the model where we have calibration results. For specific surface area. And the bottom is the validation. So it's actually a very promising model. Again, it's a very big variability we are dealing with. So it's with different mineralogy. And we actually consider it very promising. The next parameter in this model that this project that we're model was contaminant binding. And it was on a field scale, but we have actually also one on a bigger scale. Examples that Cecilia Emerson has published from New Zealand. And we have one in there was just sent for revision, which is on a national scale. Here we have been looking into the sorption of two different contaminants. And I mean, sorption in addition to degradation and then transport controls the mobility of contaminants in the soil. So a very important property to know. And this is also it's governed by other properties and includes texture, organic matter, oxides, for example. And here is the model for KD, so sorption coefficient for glyphosate that you know from Roundup. It's a known herbicide. And you can see the seeds from the reference and the predicted plot. It has a very nice correlation. You can ask why we were able to estimate it with spectroscopy. I mean, it's not spectrally active property, but it binds strongly to iron and aluminum oxides and phosphorus. Phosphorus not spectrally active, but the other two are. And that's why we're able to establish this good correlation. The other contaminant was phenentry. And it's sorption coefficient. It's a PAH that originates from combustion. And we got even better correlation for this property. And that's due to the fact that it binds strongly to organic matter, which is strongly related to different regions in NIR and actually also visible region of the spectrum. And this is another project where we were using spectroscopy for the estimation of engineering properties. And these were up-and-back limits. You have a small figure at the top here showing you how the from mechanical behavior is changing or how the consistency limits are changing with increased water content. In this study, we have looked into plastic limits. So where the soil is changing from semi-solid to plastic phase and then liquid limits. So when it becomes liquid and also plasticity index, which is simply the difference of the two. And again, a big data set, actually international data set. The top plots are calibration plots for liquid limits, plastic limit ends, plasticity index, good correlation, maybe below us, you could say here for the PI. But the validation also shows actually a lot of promising results where we can still establish a good correlation and we're able to estimate these properties good and actually not significantly different than if we would do them using PETA transfer functions. And finally, I think I have to speed up actually. I would like to mention our field applications. Since I showed you the sensor, we also work in the field. I talked about how the sensor is built. So we used this platform to map two fields. We've been driving around. I've seen those trim lines here very dense. And what appears for you most likely like lines is actually points where we collected spectral and easy data and temperature data. And after we have collected all this data, we use spectral information to be compressed. We use principal component analysis and then clustered the data to select representative samples for calibration purposes. So the representative samples are the black triangles where we went to collect samples. Then we send them to the laboratory for wet chemistry analysis. And yes, we still need wet chemistry analysis. We cannot go around. Once we got the results, we established calibration models for carbon and texture. And these models were used to predict all the remaining points that we have gotten through the both fields. Once we had those prediction points, we mapped them into those prediction maps of surrogate carbon, clay content, silver and sand. So that's just one of the examples how spectroscopy can be used in the field. But of course, there are a lot of challenges. We have very unstable conditions as opposed to how it was in the field in the lab. And we have moisture content, which is actually probably one of the biggest problems. And in the same line, I would like to tell you a little bit about the project that we have initiated in November last year. It's a European joint program on soil where we got funded a project called Probefield with a very long title, which I'm not going to read. But this is a project led by SLU and Bo Steenberg, who was also giving a webinar, I think at the very first webinar. I'm co-leading. And then we have 14 partners and 12 countries. And you can see here on the map, the countries that are involved and because they represent the ability of spectral libraries from those countries. For the main aim is to do measurements with soil spectroscopy in the field best as good as possible. So first we are developing methodology in the field to collect spectra, which are as little affected as possible by moisture. Then we're applying different methods, mathematical methods to remove moisture from the spectra and also try to combine the spectra with spectral libraries we have developed because these are actually developed on dry soils. So there's a lot of fine-tuning and figuring how we can apply these to field measurements. There will be also some work on sensor fusion and 3D field estimation as well. So we are expecting to have a final protocol of the best practice for field application. And now I would like to finish with a few words about our activities. So we're trying to put spectroscopy in all sort of soil courses. We have at least a little bit so the students get acquainted. But we also have a dedicated course for soil spectroscopy. It's a PhD course and it's both a combination of theoretical lectures and a lot of hands-on experience. So the students get to use the sensors that we have, they collect their own data, then they work with the data on the computers to do some both qualitative and quantitative analysis. So we first time organized it last year and we have external lecturers here, Bostinberg, Njohana Bettening from SLU. Then we also took some lectures ourselves with Cecilia and then a colleague Anna Smula. And we had a guest lecturer, I.R. Bendor, who introduced the students to remote sensing. So as I said, this is the list of all the publications that I have been referring to with showing the results. But I would like to finish with this slide and acknowledge the spectroscopy team. It's all the students, PhDs, masters, postdocs that have been involved in the spectroscopy work since I started back in 2008. And I didn't get to show all the work, but here you have the names and if you if you are interested to hear more stories about how we use spectroscopy, you can google their names and find some more applications. So with this, I would like to thank you very much for your attention. And if you have any questions, please put them in the QI section. Thank you. Thank you, Maria. Very nice presentation. And I would suggest you one slide with all these publications so you can share it in the chat box because I have some colleagues who are asking the publication or etc. And the second is that you can see I choose a couple of questions to answer live or later so you can read a little bit and then later we can bring into the discussion. Yes, sounds good. Yes, thank you. Yes. So now I would like to pass the floor to the Dr. Nikolaj Bok to introduce the spectroscopy development from the manufacturer point of view. Please, floor is yours. Thank you. I will share my presentation immediately. I hope you can see it. So, as you mentioned, Maria and Yi and Fos have a long history together and it has been a journey, a very educational journey, I think for both university and this instrument manufacturer. And I think that's an interesting story that might inspire some similar projects out in the world. As mentioned, the history of Fos and all universities is about 10 years old and it's stronger than ever the collaboration so it's always a pleasure to be able to make a joint project and also to share this story. This particular presentation, since Maria has been so kind to present, I would say, the technical findings which our instrument is based on, I have taken the perspective to explain how a commercial company, an instrument manufacturer looks at soil spectroscopy, what we see, what kind of potential we see in the market and why we are interested in the first place. I think I have an outline. First and foremost, I would like to introduce Fos to those who doesn't know it, then talk a bit about the market and what we can do for soil testing and then to summarize in the end. So, Fos is a company founded in 1956 by this guy into the left Niels Fos and we have throughout our more than 60 year old history been producing instruments for measuring various properties related to food and agriculture. We have a whole range of different instruments as you can see, they have evolved a little bit over time and but I think that they are clearly among the cutting edge instruments of the industry. We have both benchtop instruments, as you can see most of them here, but we also have instruments that belong in a process environment for manufacturing milk and grain products and all sorts of things. It is by all means high tech instruments we produce based on a couple of core technologies and I think that without a doubt spectroscopy is the core technology. Near infrared NIR is probably our largest strength but we also have several instruments using the mid-infrared or FTIR range and we even have X-ray instruments in the field. Any good instrument is not just a piece of measurement technology, it comes packaged so that the user can actually interpret the data easily. So, we use a lot of additional technologies to make the data easily accessible to the end user. As mentioned, it is food and agriculture that is the heart of our products and there is a range of different, shall we say, what we call customer segments, milk, wine, meat, feed and grain and then we have the shall we say a unifying customer segment which is a laboratory, a larger research or commercial laboratory doing tests and this is actually where the soil activities in FOS are located since the soil testing generally speaking are conducted in larger research or commercial laboratories. Briefly FOS is a family owned company in Denmark but we have a global presence with most of our turnover outside Denmark about 300 million euros last year was a turnover and I think we are around 1500 employees so we are by no means a small company although there are bigger, we are a very well consolidated company and as can be seen here we have a global presence with FOS companies in most of the world and FOS distributors covering the rest of the world. So wherever you are in the world it will be possible to both buy and service a FOS instrument. Going on to how we see the soil testing market from shall we say both a combined technical and commercial viewpoint. The motivation, I think we share that with everybody else, there is increasing population and of course the use of and we all need food so we need to to manage our soil well to be able to produce enough food and the good old saying if you cannot measure it you cannot manage it it certainly applies to soil as well. The problem for some is that the very many soil analysis methods they are slow and labor intensive and they use a lot of chemicals and they are most of them actually from the first half of the 20th century although it should be said that several are actually quite a bit older than that so while there is a pressing need to analyze soil then the methods to analyze them are actually quite outdated. That is of course an opportunity for a company as FOS where we want to go in and offer a better solution so this is both a challenge and an opportunity as we see it. The development of the soil testing market follows the actually the trend in food production and in population there's an we see an annual growth of about 10-15 percent that is not exactly matched with the growth in the number of laboratories because many laboratories they have a tendency to consolidate simply because larger testing facilities are more efficient than smaller ones. But this increase in soil testing is driven by of course the desire for soil management to increase the yield but there are also legislation that in many countries actually it is mandatory to perform soil analysis in order to be allowed to to fertilize the field at all and also there are a couple of very interesting mega trends that are pushing on this development as well. So all in all it is very interesting for a company like FOS to enter the soil testing market although it is not where we've had our main focus until about a decade ago. Touching a little bit of these mega trends I found them quite interesting. The concept of precision farming has gained more and more traction over the last few decades simply due to the perspective that a real soil map or the real soil might look like something in the lower left corner where but that is for most practical purposes inaccessible to the farmer because that many soil analysis are not available rather they have to settle for something that looks like something in the upper right corner with a much lower precision and a much poorer understanding of the soil. Of course more testing will give a more precise understanding of the soil and a better utilization of the soil. I've put some company logos on projects that are ongoing not all related to FOS but just to give an impression that these are some really large players looking into this. A more recent but perhaps even more pressing mega trend is the concept of CO2 sequestration in soil. It's quite well known I should say that there's too much CO2 in the atmosphere and also as Maria mentioned there is actually too little carbon in the soil so the idea to take some of the excess carbon from the atmosphere and put it into the soil is seen as a win-win concept and I really believe it is there will be less global warming and there'll be a healthier soil and for instance Microsoft has put in I believe it was around a billion dollars to actually pay farmers to put in the carbon into the soil and thereby take it out of the atmosphere but of course there's a large documentation requirement in order to document that this carbon is actually into the soil and simply managing this vast amount of soil analysis for carbon is a real challenge using the existing methods. So what are those one might ask if you're not familiar with it a normal soil is taken from dock out of the field and transported to a laboratory where it is dock it is registered. Then there is a sample preparation step which is generic for all soils and through all of the world with very little differences and that is to dry the soil using normally just a conventional oven overnight and then to sieve it to get rid of rocks and insects and leaves and other particles that are per that definition not part of the soil. Then the sample is split into several fractions where the individual analysis then is conducted using some specific additional soil preparation method but this drying and sieving is a generic soil preparation method that is and then that's an important point for soil spectroscopy because that is actually the only sample preparation that is needed. I'll come back to that. Then the results are generated in a report and sent back to the farm. So this is a quite it can often take several days even a week if you if you require some specific samples so it is not by no means a fast process. The report here's a soil report this is a bit flashy one with some nice colors but the point is that there is a soil chemistry and soil physics and soil biology part often and the spectroscopy can help with not all of these but in particular with the soil physical parameters. To the left is a little bit less flashy but perhaps more normal a report where you have just a huge amount of parameters that can be analyzed and most of them cannot be analyzed with spectroscopy but and these parameters they are site dependent and crop dependent and season dependent and the spectroscopy can take as mentioned a few of them but some very central parameters. The soil organic carbon is probably the most interesting one from our perspective but we have also the clay, silt and sand that's the texture class being a well documented and well known parameter that that spectroscopy can analyze and furthermore also a parameter such as pH and CEC being the cation exchange capacity. There can also be several others that can be analyzed here I just took a few key ones and to the left here in this table are some accuracies that I will not dig further into I will not talk so to say specifics but I will come back to the accuracy what is the expected accuracy of soil testing compared to conventional testing because that's actually a very interesting but not easy question. So given that this is how it looks today what can FOSS offer and the main product that we have to offer this is our top line spectrometer it is the DS3 which is the newest generation of our DS instruments and also the instrument type that Maria has been using with good results and it has a wavelength range the complete wavelength range from from including the visual up to the all of the near infrared which gives a lot of a lot of options in many cases one might be able to use a narrow wavelength range but this entire wavelength range that certainly gives all the options that one can imagine within a near spectroscopy. Another unique feature I dare say is that our instruments they are standardized to an extent that nobody no no other instrument in this world they are identical our instruments so comparing results from one instrument in one part of the world to another instrument in another part of the world requires absolutely no spectral adjustment modification standardization anything they are standardized already so for instance in a concept of academia sharing data that is just sending the data back and forth then it is completely accessible as is furthermore it's a it is IP65 which can be nice in an industrial in a in a soil laboratory which can be a bit dusty it is as any spectrometer fast and easy to use once it has yeah once once you understand how to use it and finally FOS I think we have we have some some unique digital services and calibration software so that when you have all these soil spectra and want to do something with them our software is I think the easiest to use in the market. I have a small video which I cannot for technical reasons present from my computer but it will be shown after my presentation here just demonstrating the instrument house used but as mentioned an instrument is not a product in FOS from FOS point of view you need to have of course an instrument it's a very important thing but also the digital services it is so you can manage the data you can develop the calibration and you can share data there is a lot of the training we have several training videos how to use FOS equipment we have developed for soil in especially we have developed an application guide which is a document on around 15 pages or so which is designed to easily get an inexperienced user going with respect to both spectra soil test soil spectroscopy and calibration and then finally we have a a startup data which is for the user who doesn't have access as Maria does and probably many others also to vast amounts of spectra and reference data so we have something to get a new a new she would say project started which is not global data as such but but it is certainly something that can help in getting some quicker results and so I will just end my my presentation here with some example which we have with a on on on so to say a practical example from a major soil lab I just have one more on on on demonstrating the so to say both the quality of the calibration but also the usability of a of soil spectroscopy and that was this was a commercial laboratory I can say that who were used to do it in the old-fashioned way with the combustion methods and with the chemical oxidation but they wanted to have something that was easier and with the fewer chemicals but they were of course as anybody else cautious well could this actually produce results that were good enough for their purpose and the main parameter as mentioned was soil organic carbon and cutting over all of these I will say with the all the spectra and this that going straight to the to the result we found that there was an error of course on both on the NIR we could document that which was on the order of 0.4 each percent absolute on the soil organic carbon level and this was soil organic carbon typically in the range between 0 to 10 percent and we were of course curious to hear the lab well what do you think about this is this a good result to you do you need higher accuracy or is this is this something that that you can use for your customers and they were actually very happy with it because they knew from their own internal internal and external ring test that their own error was on the order of 0.31 and of course using the instrument result as a calibration the calibration the basis for the calibrations only shall we say adding 0.08 percent to the error was very impressive um so um of course there were some outliers to the data but it was interesting to to um to for us to investigate well well where is the error an outlier that is if you have a a a a spectroscopy prediction that says something that is completely different from the reference result and then we analyze them with normally that is not so easy for a commercial laboratory but we were so fortunate that they were able to retrieve the samples and reanalyze them using the reference method and we actually found that in all of the cases we analyzed it was the reference value that were incorrect um and um that was uh I wouldn't say it was a shock but it was certainly something that that was uh thought provoking that um that it is a completely normal thing in a laboratory where you analyze thousands of samples to have errors on some of the samples and that can be weighing or it can be something to do with the data transfer um but it can also certainly be to the fact that you have to down sample the method so uh and down sample being uh that you have you have your your sample coming from the field which might be a kilo of soil or something like that where you have to extract only about one gram of for the reference methods but where for the spectroscopy methods as as can be seen in the video you use a much larger mass of soil several hundred grams so this down sampling effect is really a major source of error we believe speaking a little bit about the origin of error source of error it is it should be noted that um while the reference method is shall we say per definition correct because that is the reference method um then uh there and comparing that the NIR or any spectroscopy method for that matter is a secondary method and there is by definition an additional a source of error there i've explained before that it that it was in this case only a marginal source of error so it was not shall we say something um critical then we have for the sample preparation um most most spectroscopy method methods and but in particular near infrared spectroscopy requires no additional sample preparation beyond the standard sample preparation of the laboratory being drying and sieving typically at two millimeters no no other sample preparation is needed you have uh you have that um the sample prepared in that way you simply scoop it into the cup and you measure it and you're done in one minute no more uh many reference methods and certainly for for soil organic carbon requires some quite substantial down sampling and subsequent chemical chemical oxidation or or shall we say handling in some way um which can be a major source of error and and that also perhaps i should mention here because i actually saw that it was a question on that that also goes for uh mid infrared spectroscopy where the sample amount needs to be really really small so so so that doesn't relieve the sample prep source of error here NIR spectroscopy is really very unique but finally when we are in in in shall we say in projects where you don't have access to many repetitions of the very same uh field then the sampling is the by far largest source of error you have a field which might be one hectare uh which might be 100 hectare it might be one hectare um or much larger and and you can only take one or two samples from that uh from that piece of land and obviously recalling some of the pictures i shown there are a lot of variation there that you simply do not capture using shall we say the or given the constraints that that normally exist in in a normal soil sampling methodology so i think that's worse to consider that when we talk whether or not a little bit more or less error comes from the reference or the NIR actually the largest source of error is from the sampling perhaps only a few academic studies solving this but certainly for a production laboratory and for farmers there is no way to handle the soil variability in any way so that this sampling is not the major source of error that's important to recognize so um summarizing um from from Foss's point of view we see that NIR spectroscopy for soil is fast and easy to use and cost effective there is a minimum sample preparation uh no sample preparation actually beyond the standard which is conducted which is sort of the basis for all other uh tests anyway uh we have tried in Foss and succeeded i believe to make uh to make it as to make it very easy to get going with database management and calibration development and for this particular study it was quite interesting to see that that actually the outliers is a problem that that that simply can be if not eliminated then greatly reduced by the soil spectroscopy probably due to the down sampling problem i didn't get around to talking about the repeatability um so i will skip that point so in conclusion we find that NIR is a very suitable technology for handling large amount of soil samples with a high variance and with that i would like to thank you all for your attention thank you very much uh Nicola and i will i will try the video yes please and that video is uh shall we say a demonstration video capturing shall we say the product how to use it um it is just a few minutes long so it's not a an in-depth video in the meantime Nicola i think there is a couple of questions in the qa specifically for you or for the Foss you can take a look my name is Nicola Ibog i'm a soil application specialist at Foss today i will talk about analysis of agricultural soils using near infrared spectroscopy in particular with our DS2500 analyzer the soil application package consists of a number of elements an instrument which will analyze the soil sample cups containing the actual sample an application guide which contains all of the elements that are needed in order to develop a calibration as well as a number of digital services and solutions for data management the actual measurement of the soil is extremely simple you open the lid take the sample cup and place it close the lid and press measure on the instrument while soil management and fertilization is becoming ever more data driven the analysis methods themselves have actually not involved for a very long time and while these methods are well recognized and established there are several drawbacks near infrared spectroscopy requires minimum sample preparation and no use of chemicals this particular sample is being analyzed for its content of organic material termed soil organic carbon soil organic carbon is a very important parameter and other parameters which can be analyzed includes texture ph cation exchange coefficient and total nitrogen as any other soils entering a soil testing lab they have been dried overnight at slightly elevated temperatures as well as sieved at around two millimeters to remove branches rocks animals and other larger particles the result is a fine powder which is very well suitable for nir analysis in conclusion near infrared spectroscopy is the soil analytics methods of tomorrow it produces vast amounts of data it is easy fast and minimizes errors from manual handling thank you very much julia for sharing this video and thanks and thanks for preparing this video i think the another thing is i think i just because of this video i think some of the colleagues in the chat were asking the price of the instrument etc and i would suggest colleagues if you would like to know more about the company if you feel free to get in touch with a first company or you are still free to reach me i will be happy to make a bridge make a link between the countries and the manufacturers not only the first of course other manufacturers as well i i don't think that today is the is the proper moment to discuss the price of the instrument and because we will have a lot of interesting discussion uh be followed and it's been always a stressful due to during the presentation period and thanks for sassin sassini had a lot of questions and i can see there is still a lot of questions to answer um we choose some questions to answer alive as a discussion and also please also feel free to write your questions in a qo and a box um in a meantime nicola you're also please feel free to some of the questions probably you can type to answer the questions and the first question actually is talking about uh one colleague asking some researchers suggest the m ir over n i r for organic matter clay estimation the same has been mentioned in the training material sent out by the previous set of webinar how has been your experience within the same and i would like to address this question because uh that training material was developed by us in a glossary in a field and uh indeed that training material was uh was a really beginner for a really beginner level of uh uh colleagues to learn this technology so we introduced the both visible near-infrared and m m ir range so we uh from our experience and existing data and study we saw mid-infrared reflectance spectroscopy in the soil application will have some advantages uh intensively predicted the the virus of the soil properties uh and also the accuracies has has some advantages compared to the near-infrared spectroscopy but also please also be noticed we also mentioned that each type of the spectroscopy each range of spectroscopy has their own advantage and disadvantages so we don't really recommend the which range of the spectroscopy you are going to use and if you have enough resources we would encourage you to have to equip both range because each of them have their advantage and advantages and if you don't have enough resources you have to make a decision depending on your budget and your objectives so that would be my recommendation it's very really difficult to say which one is really better than another one so uh can I supplement here yeah yeah I totally agree with with your answer here I just wanted to link also to for example the aims which are very important this is where you should really be clear with yourself what do you want to use it for and I showed some examples of using NIR in the field and if you would like to take an MIR sensor in the field it would be even more prone to the differences in soil moisture and already NIR has has its limitation and we have to uh pretreat the spectra to remove moisture effect uh mid-infrared range would be even more affected and even though there are some instrumentations available to analysis in the field it's still um highly actually uncertain so again if you would like to go for measurements in the field you would for example choose an NIR so the first question to answer would be what would be the the final application of the sensor you want to choose or use yes thanks all right there is one question from Fennie I think it's quite interesting to ask that on the model because I would like to read this question because so we are recording this webinar later on other participants also can know the question the question is about the other model for macroprocity it is interesting that you report a better result with NIR than with the peter transfer function even though the prediction with NIR was not that good do you know what may have caused this a different scale of a calibration field the scale national scale or the representation of influence you saw your property in spectral yes uh thank you Fennie for this question it is actually a very good question and and I think it goes down to the fact that in peter transfer functions we are using other available salt properties in this case for macroprocity we used silt sand and organic carbon contents so these are this is only information about the contents that ptfs are concerning and in in case of NIR spectroscopy it's not only the contents but it's also they are also reflecting the qualitative information of of these properties so it comes to clay for example they would also reflect the mineralogy as we have distinct spectral responses for different minerals or the same force or getting matter I think that it's it's much more to it when we look into spectra because it's not only the contents but it can also tell us what's the quality and that's most likely what happened here so we have somehow much more information in the spectra than just the contents that are used to to develop peter transfer functions thank you Maria the next question is Cecily you would like to answer in live about will this course that was mentioned that be organized this year yes maybe I can with the question to Maria she knows more about that yes of course so we we actually decided to do this course every second year so it's not gonna take place this year but we are aiming to do it in the following year and we can also maybe put a link to the course website in the chat I have it somewhere here so I'll send it to you in a moment but yes we'll we'll organize it and it's and it's a PhD course but it doesn't have to be you don't have to be a PhD really to attend it we had actually some master students as well we even had some well it was one lady from a conventional wet chemistry lab where they are interested in introducing spectroscopy so it's actually different types of participants that's that we can accommodate in such a course so I'll send you a link in the moment in the chat yeah in the meantime I also would like to address close on spec is well connected with all the institutions and all the research groups once there is any activity related to the training or or courses we will share the information so please regularly visit our website or facebook group etc and also I believe force is we have we are discussing with the force to see if there is any potential to develop some training program or so to share some protocol because just now has a one question I saw they showed the develop some protocol so I think a force also has a SOP for using the instrument to measure the soil all this I think we need a collective effort to bring this knowledge transfer the knowledge the expert knowledge to the society so we still have a lot of work to do in this regard and another question is Maria could you explain more about how you did the NIR soil classification map you selected some web band for this or how you did yes thank you for for this question and no for this exercise we have used the full visible NIR range so what we did we have collected spectra of all the samples that we wanted to include in the mapping and we have performed principal component analysis on the full spectrum and so what the principal component does it decomposes this huge matrix into several or fewer principal components and this is what we have mapped so we have mapped the output of principal components and these were the PC scores from the first three pieces so it's it's not it's not the spectra as such but the actual PC scores this I hope it answered the question I can supplement a little bit I would like to I would suggest the colleagues to watch a video recording I think the webinar three last year from Professor Alex McRanny University of Sydney we had a dedicated discussion during his webinar and I kind of agree with his opinion for the future application if we want to use a spectroscopy technology to facilitate the soil classification one of the approach is because when we do the soil classification when we classify a soil profile we need the soil property data of the soil profile and in order to do that traditionally we send the soil samples to the wet chemistry lab to do the wet chemistry analysis which is a very time consuming and expensive so if we can apply the soil spectroscopy technology to estimate the soil property different soil properties in the fast and the costly efficient way that we or and then we we can use this result to classify the soil profiles and then use a classified the soil profile information for the mapping soil classification mapping purposes that's also another advantage we can take from this technology okay Nikola Nikola Nikola and Maria any prospect for soil spectroscopy in Africa well I can certainly say that there are several abortures in Africa doing soil spectroscopy and I believe that in Kenya you probably know that Maria there is a a an important center for soil spectroscopy coordinating the whole African region but perhaps you can tell a little bit about that um do you mean the apsis sorry now what is it called yes I was yes yes yes that's true I mean there there has been actually a lot of activities when it comes to african spectroscopy and I think it's actually mainly made in for a range I'm not aware of actually activities with a near inferred spectroscopy and african soils but as far as I know it's jubia it's jubia they are doing both in there and okay yeah the the study that I was referring to at the end of my presentation was actually from Africa yeah yeah there are some countries I do of course definitely both at the UN research institute and private sector which put more focus on Africa because definitely need more resources to help them develop such technology and one quick question for for Nikola and also for the not question actually probably is a message delivered to force one participants ask can we do online course from the force so probably it's a message to deliver to force if a force can organize some online training maybe together with Maria and the gloss alone we can maybe we can develop something like that please deliver this message to the force yeah I mean we have we have demonstration videos and and stuff like that but it's not really a course as such they are more like demonstrations but I think that it's perhaps a good idea to to to make a course also in the same shall we say with the with the phc course perhaps a couple that in some way yes definitely I think it it would be very nice to to combine it actually and add also your knowledge and information instrumentation to such course definitely um can I can I actually follow up on this PC classification map because there are so many questions coming I'm getting a little bit of stress the piece and so just to follow up I it seems like it was a very interesting exercise that I showed so the question is which sort properties or property does the pc represent and so again these were the pc scores and the only way to know what property they are reflecting on what variability they are explaining is by looking into the loadings so it goes a little bit to the basics of of principle component analysis and I have showed a plot with loadings for those three pieces and it was clear that for example first loading had high information related to organic matter content the second one to play content for example so this is the way how you can you can read this sort of hidden information so you look into this course and you have to also correspond them to respective loadings in order to make sense out of it yeah so that was just to follow up on the classification map yeah yeah um I don't know if Maria did you because there's a question constantly coming in I don't know if there is any of probably we cannot answer all of them I don't know if there is any particularly you feel interested to yes I'm also getting a little bit confused actually so just give me a second I would try to pick up with one more there is one a colleague asking to share the share the training material the link I'm sharing to the chat box now so colleagues please feel free to download now I'm a little bit confused if we did talk about this there is there is one comment on the need of protocols for laboratory measurements can you please just read the can you please just read the yes of course it's uh we're organizers we need standard protocols for proximal in in laboratory methods how can we have training course for developing these protocols and experiences for practical I guess application um yeah I think that this is a very good comment and there is already a lot happening when it comes to standard protocols and the measurements done with NIR spectroscopy and and soils and there's an initiative called IEEE and initiated by IRL Pandor from Tel Aviv University then there are several um colleagues um some spectroscopies working there are several work packages that are addressing the issue of of more standardized way of of doing the measurements and we are facing a situation there are already quite quite many countries um doing the measurements and everybody has their own protocol on how to do it and of course if you would like to in the future share the data it would be very useful to have it done in the standardized way so I guess that um yeah this comment is yeah it's actually very relevant and then very much on what is happening right now and the the different initiatives um within the the protocols and and needs for internal standards and similar approaches yeah there is another thing I would like to make a supplement is it uh Glosselon it's actually the Glosselon one of the main Glosselon's mission is to develop this standard protocol for the measurement however we have we we are facing some challenges um about the some FAO rules we are not allowed to mention the uh mention any commercial brand to our any of our publications uh so it's quite challenging for us to publish such documents related to the protocol etc but we are working on it and hopefully we will soon to uh to figure this out and release some protocol out and get this knowledge and experience to the countries uh one question actually uh Maria because one question in the beginning I don't want to read because it's quite long but he asked how do you see the progress in making soil spectroscopy and a soil spectral library used available for application probably it's a it's a very big question uh globally it's difficult to answer that will take days to discussion probably you can you or Nicola can just briefly um say uh express something from your perspective for which direction you want to go for the Danish application in five or ten years uh go what is your in your vision what will happen in a soil spectroscopy application in Denmark at least um yes Nicola do you want to start um well um first first uh perspective we have we have certainly have uh uh shall we say the the actions that we and the research project we engage with that is they are all have an all they all have an overall global perspective we are very much happy for for this uh collaboration we have with old university to uh to where we will try new things and we are really innovative in order to to then to market them to a broader audience so I really hope that we will be able to to still get some good ideas and try some some some things that are pushing the limits for what is what is possible and then our our road will be to see well has this any broader use beyond the the research world um so I will not go into any of specific parameters that Maria will but I think that uh that that there will be both new applications within this particular type of spectroscopy but also other types of spectroscopy uh we're having those shall we say in our in our portfolio developing them and I think we will see soil applications being uh we're having a broader portfolio within forces instruments yes I can supplement from from our perspective and of course I mean since we are a research institution we're a university so it's very much um uh challenging spectroscopy for for different applications but the actual application in real life is of course not really that much a part of what we're doing and this is why again going back to what you said Nicolae that it's so important that we actually establish those collaborations with private sector consultancy companies and that is what I could also recommend for others if they have a possibility to do it because this is how you can more quickly reach to the end user and this is what we would like at the end and that wouldn't only science then of course it takes really years before we can see some some results put in action and yeah by by um defining some kind of collaboration with private sector that can actually reach also the final um end user it it's it's probably the way to do um and I was actually I was a little bit not sure about the question really because um now I don't see the question it's lost somewhere uh among the others but I thought that there was also like how to make sure that we can actually apply um or sort of validate spectroscopy in order to be able to use it wasn't it I think this is what was actually the question I think maybe you can search in the answer there because I try to uh reduce the number of the questions that you hear is too many so yes okay but then just to follow up because this is like our perspective from from Denmark and Denmark is relatively a small country uh we can um relatively easy let's say uh compile a database that would be representative but for many I guess and and for the use on a global scale the question is how can we actually make sure that uh whatever data we are generating is applicable in other regions of the world so I think also the initiatives that um Glossalan and so spectroscopy group is is um representing to sort of join the forces and experience from different scientists and shared spectral libraries that's that's probably the way to go thank thank you Maria and probably we can we are getting to the end there is a question it's uh it's also a good to discuss is what would be the adequate procedure for account the uncertainty of the near-infra-spectroscopy model for long-run soil organic carbon monitoring experiment the uncertainty of the I think of the prediction result that's what he must look I can I can comment on that um the uh that is not easy very interesting from a from a for for many but not easy because the errors are coming from both the NIR being the calibrations which can be which can be identified I would say that is that is not so hard actually um performing normal chemometrics which I don't know if the if the the person asking the question is familiar with will reveal many outliers where you have a prediction saying something and then a reference value saying something else and typically you will assign that to be the force of the prediction saying that that is wrong the hard part that is to identify when you have a false reference method a reference result because that is really the case in real world if you have many thousand samples which is not uncommon or even have a many hundred of samples which is not uncommon for a a study it is simply from statistics it is likely it is most likely that one or more of them will be off by perhaps 50 or even 100% due to some typo or something like that and I think it's very important to look not only not into the to the I mean to be critical about the reference results also because when you have that that can really that can really be damaging to a calibration and the only way to do it is to be to be critical and to look into the data and to double check of course have the samples available for double checking so I think those are some of the advice I will give beyond what is chosen normal with repetitions having good statistics for every sample being thorough all these things yes if I could follow up but actually it's maybe something you can follow up on this because the whole issue of us relying on wet chemistry data which is I mean this is this is what we were we were we cannot go around in order to make calibration models we have to have input from wet chemistry and and this is also what glossolan is is about right and and the laboratory network where the main aim is to to give some guidelines and standardize the already available wet chemistry analysis so that we at least try to minimize the errors coming from from from wet chemistry so we maybe you want to follow up on that yeah I remember the Maria before we had quite a lot of discussion on this the word chemistry because I think there is one thing very important to be noticed is a soil spectroscopy is highly relight really highly relies on the word chemistry because we need the word chemistry data for the calibration and later on when we do the validation we also need what what chemistry relies on the word chemistry data to do the validation especially during the last 100 years that we are we have been relying on the word chemistry analysis for last 100 for last one or two decades so we truly believe what chemistry provides us the true value of the soil data soil samples but if but is it really true value of the soil samples or not I would suggest the colleagues go back to the laboratory do some soil analysis like a soil texture analysis or even a carbon analysis and try to use the same sample and repeat doing the wet chemistry analysis and then you will know what kind of results you will get and that time you will understand why spectroscopy estimation results from a spectroscopy brings uncertainty so that's a very I have to say it's a very important experience you you should go through it if you want to do the spectroscopy so that's what I would suggest. Nicola there is one question from Rafa Atiyah actually Rafa Atiyah is a member of our ITPS and she is from Tunisian Ministry of Agriculture and currently they are planning to develop their own capacity so within their ministry and she's asking could you assist the calibration of different instruments spectrometer because I had a discussion with them they would they really would like to have technical support to build their lab or install the instrument and train their staff for the measurement from a sample preparation calibration instrument export the data use the data management etc I think it's a whole package we're talking of course it's not something we can do it tomorrow but probably it's another message you deliver to us as well. Yeah but in general our procedure for selling and providing instruments we don't we don't see ourselves as an instrument provider only we are solution providers so when we provide an instrument which is of course important then there is a whole package of whatever is needed for the customer to to be able to use it actually and there are these things that are mentioned some documentation this and that but it is also possible to that will have to be arranged with the with the local force representative to her to have assistance either either in person on site or online in some way to to enable the user to actually gain the most of the instrument I cannot go into the detail of this specific case but it is something that I would say in general yes but but but they will have to talk to their force representative for that. Thank you Nicola. Another question I would like to response is from Omar he asked regarding the soil organic carbon do you think this methodology will be able to detect the temporal change of the soil organic carbon this is something about the soil organic carbon substation potential and I would like to mention a little bit because I'm working on both the soil spectroscopy and the soil mapping part and in global soil partnership since last two years ago we initiated a global project called global soil organic carbon substation mapping project so we are enabling countries to map their soil organic carbon substation potential cross from the past to the future based on the different scenarios based on different scenario is business as usual and the sustainable soil management one two three three different scenarios so based on the different scenarios you will be able to see what kind of land management based on for the agriculture what kind of land management you will effect in the future your soil soil organic carbon substation potential with also uncertainty assessment but what a spectroscopy can play a role in this topic because if you want to evaluate if you want to estimate the future circuit street carbon substation potential in the future then one of the most important input layer for the carbon modeling carbon substation potential modeling is the soil organic carbon map and the clay map and those maps requires a substantial amount of the soil profile data to generate the map so on that time if you want to huge amount of soil data to generate the such soil map at the national scale and the soil spectroscopy can definitely play an important role for the mapping purposes so that is my comment so you are if you are interested for soil organic carbon substation carbon potential substation potential mapping project please also feel free to to visit our website and we have quite many activities in this regard I also added a paper that we have published on the use of NIR spectroscopy for monitoring soil organic carbon temporal changes it's a paper by Fendeng it's in some years ago I don't remember the details but I remember that yeah we showed that you could actually yeah sense the differences in in soil organic carbon between was 1980s and I think 2009 so we had two data sets measured with spectroscopy from from those two dates approximately I don't remember exactly and yeah so I send the link and please go ahead you can see some more details and probably the also let me see Maria probably it's also it's also good in that Facebook group we have a Facebook group you can put this your reference papers to to post there because we have more than 1000 people in this group and actually most of the participants also join this group it's and it's also good to popular the information from the it's of course I don't know if you would we reached the 42 we answered the 42 questions already I don't know if you if there is anything you would like to see there's maybe one last question there is a question does spectroscopy affect the chemical properties of the soil um yeah I'm not sure if I understand the questions probably if the chemical properties are affecting the spectrum spectrum measurement I guess that this was what what was meant yeah so that's um it was actually a very good webinar the first one I believe by Bo Steenberg talking about the fundamentals of spectroscopy but in general what you can see the sense with NIR spectroscopy would be the molecular bonds of yeah different types C H N H O H S H bonds so anything that's related to both organics and mineralogy in short that was the answer okay I don't know if Nikola if you have we reached almost two hours I think if I don't know if Nikola or Cecily any of you have anything to add on part of our discussion I think I've talked so much already I must have said it all okay okay thank you very much all of you for this nice webinar and I have to say it's been very active and it's also always been a very stressful part for monitoring question and but I think we managed it very well and also thanks colleagues to join this webinar and also the time it's always very difficult because this time it's always difficult for the colleagues from Asia and the Pacific and also for Latin American it's very early and but we have we have recorded the video and after editing and then we will upload it to the our website and everyone will be able to watch this again and also we will upload the all the PPT and the materials so hopefully we will have more collaboration in the future and to help the countries to develop their capacity and bring this technology to the countries to the world and facilitate the soil monitoring and sustainable management and also in the meantime advanced science so thanks again and have a nice afternoon and the rest of the day thank you very much once again for the invitation and all the participants for for great questions and discussion thank you very much yes thank you bye bye