 Good afternoon. We almost have 97 people and again today we have a talk with the Australian Embassy and it is a great pleasure for you to be here with us on the most important thing before we start. There are several things that we are recording the session and later on you can download it through YouTube and you will have a simultaneous interpretation with the panelists that we have today. I welcome Camilo Peña who is the person who's gonna be with me moderating and madam the ambassador of Australia's here with us. Mrs. Thompson, Erica Thompson. Welcome Mark Phil. Thank you for being here and Camilo let's kick off now. The title today is it is livestock management in extensive and sustainable and Camilo for everyone here it's important to tell you that new technologies where livestock sector we have a clear object which is productive productiveness and to make it easier for the farmer and we need to exploit natural resources so today they're going to give us a talk mindful of the environment and success stories in Australia. Good afternoon Camilo. Good afternoon all. Today we have a very interesting number of people so far we have 131 people in Colombia and we hope that we have a very interesting session as you said Sulaam the ambassador Erica Thompson is here with us and she's very mindful of our sessions and she wants to take advantage of all the Australian success stories that can may benefit all the Colombian livestock sectors so without further ado I would like to introduce we're going to have two presentations today the first one we are going to have a Phil Teckel who is the founder of Cycle Lab together with Pierce Scott he has more than 30 years experience in the application of aerospace technologies and also managing natural resources and his mission is to focus on something new on livestock lands in Australia to take care of farms and to make it more sustainable to introduce him we need to see all the loads of animals and that is critical for sustainable livestock management and grazing industries and we need to respond to sustainable I'm sorry but he's breaking up and this is why Phil who is here on behalf of Cybo Labs they use a combination of applications of data compiling in farms with satellite images per week simple platforms to take care of all this grazing sector and Phil's going to talk about and then after that we're going to have a we're going to have Mark Trotter who is an associate professor at the medical sciences in Queensland University which is one of the most important ones in Australia for research are about variability with a special temporal and the development I'm sorry he's breaking up before the variation for animals recession in grazing industries have transformed all the industry in Australia will they do that with extensive grazing industries like with livestock so Mark is going to tell us how these new technologies emerging technologies are changing the way that Australian farmers are monitoring their cattle for with a in a sustainable way and we know we hope that all these practices and innovations in Australia are very useful for all of you in Colombia so now I welcome our presenters Sulaam said we're going to have simultaneous interpretation so you can follow it and start your presentation thank you Buenos tardes everyone can you see my screen okay slowly sharing yes I believe we can see it now okay good morning everyone apologies I don't speak much Spanish what I'm going to be talking to that to you today about is satellite assisted forage budgeting CBO Labs was established in in early 2018 but really was on the back of about 10 years of work in the remote sensing of pastoral systems in Australia and the realisation that we really had to take some different approaches if we're going to have an impact in the grazing industry in Australia so what CBO Labs does we're a satellite remote sensing and agricultural data science company and we are providing services at the moment to about 50 million hectares of individual paddock pasture biomass predictions across Australia largely across Australia's north but also expanding into the southern pasture systems and the sorts of questions that we're asking or answering are how many kilograms of pasture do we have in our paddock or on a farm how many grazing days do I have ahead of me with or without rain what is my land condition in terms of ground cover levels and how is my ground condition changing over time we're also starting to work on soil carbon related projects there's a lot of interest in soil carbon globally and we're we're applying remote sensing to a whole range of soil carbon projects in Australia at the moment as well no doubt there are lots of satellites that have been launched in the last five years I won't go through all of these in detail but just to I suppose I've provided some links there for you to to look at later on and we don't have a data problem anymore there are literally hundreds of satellites orbiting the earth observing the earth and we can provide information on a daily to weekly basis in terms of in terms of land cover changes across the across the planet the real challenge is taking that information and then providing it in usually use easily used forms to to provide farmers with better decisions so what CBO Labs does is really focusing on the the data science looking at traditional methods of pasture biomass estimation and providing those that information to producers in an easily used form so on the left hand side here this simple diagram here we've got highly sophisticated systems for automatically processing satellite imagery so as I said earlier on we're currently imaging 50 million hectares of pastures every week we have mobile apps that we provide to farmers to collect information in their paddocks and fields and farms we have a machine learning platform that we have developed to take that data collected by farmers and then use that and the satellite information together to predict the kilograms of dry matter of pasture in any individual paddock and then to make that available into decision-making systems and software packages or spreadsheets or maps or range of forms what I thought I would do is just to quickly walk you through some examples this is a typical image that's provided to a farm this farm is about 10,000 hectares but it doesn't matter whether it's 100 hectares or 1 million hectares we can still image it on a weekly basis obviously clouds are the other things that get in the way but technically technically we can image them basically on a sort of weekly basis so what you're seeing there is a is a 10 meter resolution multi-spectral satellite image process for that property and and you can see the the vegetation you can see differences in in soil color and you can see basically down to individual trees we run a model called a fractional cover model that looks at the proportion of green photosynthetically active vegetation the non-green or the dry vegetation and also the bare ground so we can see the dynamic dynamics of how that pasture system is changing through the season and and use this information to not only measure the amount of pasture but also the general quality of the pasture and then we predict kilograms of dry matter so in this image here the brown areas are the lowest biomass and the green areas are the highest biomass we can also tell you for any individual pixel how many kilograms of dry matter per hectare is actually in the individual pixel this is just simply a web application that producers can basically have on their web browser or on a phone to to look at the variability in pasture across each individual paddock we then produce traffic light maps so again taking the complex information and turning that into simple information here is a simple traffic light map showing the highest biomass areas in green and the lowest biomass areas in in red and and then you can also see at the top there when I can click on a paddock we can estimate the average kilograms per hectare of pasture in that paddock for that given week we also integrate that into into other software packages so we work with companies such as agriWeb and those farm managed software systems integrate our data into their platforms so you can look at the individual animal data and the mob data and the and the pasture data all in one all in one platform the critical factor here though is is how to train these sophisticated machine learning models and we're reliant on our clients and our customers basically helping us with this and we have mobile apps and we can provide cattle producers in Colombia with these mobile apps and adapt them to your local conditions but the core of our business is getting our clients engaged in the in the process this example here is actually a a map of a client's property where they're collecting pasture assessment data using our mobile app which feeds directly into our machine learning platform here's just an example of the of the mobile app so to be able to take pictures to be able to do quadrats or visual estimates to identify the species that are available there and the palatability of those species so this information is is collected offline so that the app works offline with no reception and as I said earlier on this would be very very easily adapted to to a Colombian circumstance for people to collect information and do their pasture assessments using this mobile app that data then feeds directly to us so as soon as you get doesn't matter whether you're in Australia or Colombia if you collected data your your your little dots would occur on on the map here and and then we can start basically using that information to to train the satellite data to predict kilograms of dry matter in Colombia we're also using some sophisticated computer vision techniques this is about 12 months away from production but to be able to just to take a picture of a of an of a of a pasture and then automatically predict how many kilograms of dry matter is in that pasture clearly that's reliant on doing field cuts and and observations as well so it's about creating a library of information over time and then you ask the question of how accurate is what we're doing so this is an example here where we've taken independent data and and collected data on a farm sorry the farmer has actually collected this data and then you can see on the top right the sorts of relationships that we achieve when we're when we're doing calibration validation at a local level the past apologies apologies to interrupt some people are writing on the chart that the the presentation looks very small i think it's because you have selected the percent interview that's why we see next slides as well would be possible to right click and hide percent of you so sorry it's only one slide and it's bigger has that and that's a lot better thank you feel i'm sorry sorry apologies for interrupting no sorry i should have done that before i thought i actually was in screen mode in full screen mode hopefully you've been able to see the slides until now so you can see here we've got past tropical pastures there that are typically over 5000 kilograms per hectare of dry matter and and an even higher and we have the ability to predict biomass in those landscapes here's another example more recently as well taking validation data and you can see the relationship there with that scatter plot on the relationship between our field estimates and our predictions over a very very complex landscape and then nationally we are we're basically rolling this out nationally now so we have the ability to scale so where we'll be in the early 2022 we'll have a system running predicting kilograms of dry matter for every farm in Australia so again there's a signal layer that basically at a country level something like Colombia we don't have scalability issues and the ability to do this and by middle of next year we'll be looking at every single land parcel in Australia and providing information to to people that want to pass your biomass for their individual properties we can also go back in time as well so the satellites we use now to collect the information and and under under the under the predictions have only been available since 2015 the Sentinel satellites the Landsat series of satellites that go back 30 years and they're available globally we're cross calibrating our models and that's going to allow us to run our models back in time and then look at the pasture production basically over the last 30 years this has major implications on our ability to predict saw carbon based on grazing systems and changes in grazing systems over time so this will be an absolute game changer just to give you some sort of view of the transportability we're doing trials in North America at the moment so you can see the little dot there in North America we've got some trials starting in North America we've got some other trials in in in Africa and France and and also Brazil now we're doing some some work there still very early days we're but we're working gradually to to take our technology to other countries and clearly very keen to look at look at Cambian opportunities now what I've just described there is it's basically around our our forage budgeting system I also wanted to just show you briefly some of the new technology that we're bringing to bear as well so recently we've become a partner to Ceres tags the world's first commercially available direct to satellite ear tag for cattle and we have the ability now to track those cattle on a three to four hour basis around the paddock and you can see here the date of this image I just screenshot here two days ago and I'm looking at basically an individual animal at 7.33 in the morning on a property about 2000 kilometers away from where I am right now and and we're looking at a heat map there of the movements of those cattle over the last over the last 24 hours we can also then look at individual animals and look at the activity of individual animals so I'm looking there at one specific animal it's actually had a high activity a medium activity over the last 24 hours and we can see where the animal was actually moved and you know this is going to be fundamental for understanding you know where animals are moving in the landscape and how much time they're spending across individual areas within a paddock as well as they're ultimately their health and Mark's going to talk about that I'm sure but the real game changer for us not only is the ability to integrate that data in terms of specific applications around animal theft and animal health but what we can also do now and this is an example here of one of our clients where we have our pasture biomass predictions that I've been showing you earlier and bringing you know pasture biomass predictions and the animal movements together so we'll be able to look at the pasture utilization across a paddock and start making predictions and estimates of how how how animals are utilizing the pasture across a paddock and and what we can do from a management perspective or an infrastructure perspective like fences and water to improve the utilization of individual paddocks and just finally I just wanted to quickly just talk about land condition and long-term land condition we also have an application in Australia which is taking 30 years of satellite data so we can actually process data for every pretty much every farmer in the world for the last 30 years and we can look at changes in vegetation and changes in ground cover over those properties and what we're doing here as part of soil carbon projects is and land condition projects is benchmarking properties using time series satellite data and and to I suppose there's a lot of complicated graphs in this slide here but on the bottom left we're looking at changes in tree cover over time and very soon we'll be reporting directly on on woody vegetation above and below ground carbon and generating reports for for individual properties on carbon sequestration and on the top right is basically a comparison where we're comparing an individual property to the neighbouring properties in terms of its ground ground cover over time and by doing this we can then start to benchmark individual properties in terms of their management practices and on the bottom right is an example there where we're taking that time series data and then ranking that property to the neighbouring properties and you can see there that that property basically was in the 50th percentile so it was basically about the same or below the neighbours in terms of its ground cover management up until about 2010 and then and then in then after 2010 they've then they've now moved up into the top 25 percent and this property here has changed from cropping practices to rotational grazing practices so this is a this is also going to be influencing soil carbon production over time and we're using these tools for registering soil carbon projects in Australia at the moment now just finally also wanted to just to touch on one of the biggest barriers to technology adoption we see all of this amazing technology the satellites the mobile apps mark's going to talk about sensors and IOT the incredible amount of technology available but it all comes back to the lowest common denominator and at the moment we have a very small number of farms in Australia that have a very high quality digital farm map and if there was one thing that I could basically suggest in terms of trying to enable the technology adoption you've got to start with some of the simple things first and without a digital farm map we really can't fully fully realise the benefits of the other amazing technology so we've built some systems here in Australia to map property infrastructure or farm infrastructure fences waters etc and do property plans and again this is global technology and could be easily adopted to to Colombian applications but you really need to have this farm map before you can take full advantage of all the other technologies that I've talked about here today okay I might leave it there um so probably my sort of 15 minutes is up um I've provided a few sort of additional slides I haven't worked through but certainly it's a very exciting times and you know we we're looking at a range of technologies that are really supporting forage budgeting in Australia as one example uh one of our key clients recently a very large cattle producer in Australia they were able to bring forward they have six and a half million hectares of land they have about 400,000 head of cattle across the operation about 20 farms and they were able to bring forward their forage budgeting by two and a half months so rather than doing their forage budgeting in July when a lot of the animal movements have already been made they're now making decisions basically in April on our information to to to drive their animal movements which have a range of implications in terms of risk management costs of transport animal welfare and a whole range of things by being able to make those decisions early and then it's through you know applications like seris tags that we're really taking it to the next level now looking at real-time real-time animal movements and pasture pasture budgeting going forward so thank you for your time and we'll look forward to discussing our sort of expansion of these services to uh to South America we're certainly um now working on a range of fronts in the global sort of applications at the moment so look forward to working with you in the future thank you field that was a very interesting presentation that was heard by 306 people who are joining us right now great audience and people who are from all over Colombia and not only from Colombia but also from Mexico uh people are tuning in from there and thank you so much for being with us this afternoon and we thank those farmers who are here from different parts of the world and in terms of Colombia the people from Cordova uh in the north of the country in Guajira and we also thank you for being hard people from from Elgar from Sucre and thank you so much for being uh with us today and from Rio Hacha the local project and thank you so much for being with us uh in La Mojada which is a very interesting part of Colombia and very productive thank you so much uh for being with us Camilo uh you have the floor um and uh we have more than 300 people with us today okay thank you so long so now we welcome uh Mark Trotter and as we said at the beginning he's gonna tell us about uh the precision livestock management has uh influenced the industry of crops uh in and the extensive grazing industry so I give you a mark so he'll give he start with his presentation screen oh let's see how that goes can you guys see that screen okay perfect Mark excellent right oh so i'm going to be talking today about some other technologies that we're looking at um applying Australian extensive grazing systems um little bit of oh hang on let me get this working a little bit of background on me first so I think it's always nice to know who's who you're talking to I grew up on a dairy farm and a commercial beef operation on the mid-north coast of New South Wales so that's in southern Australia and then I went through university in that state and worked in the um agribusiness industry for a while and then I've just in the last five years moved up to to northern Australia to seek the university which is based in Rockamp and which is a more of a tropical um type environment and so in some ways similar to to Columbia in a lot of ways in terms of environment I'll be talking today about some work that I've been doing but there's a whole bunch of other contributors involved in this but in some of the work that we've been doing I just wanted to acknowledge those folks um I'll talk today about two key technologies the first one I'm only going to briefly touch on this one is our walk-over-way systems and then I'm going to focus on some of the on-animal sensing work that we're doing and then Phil talked a little bit about that earlier with Sarah's tags just to start with though the walk-over-way system this basically is essentially having a set of scales out in a paddock and the animals come into the uh into the paddock or come go into the water point through a set of of spear what spears so a one-way gate and they walk across this um uh weighing platform and then that data is collected and sent back uh uh by a radio link and so essentially what we end up getting is a live weight on a daily basis of each individual animal in that particular paddock and so traditionally we would have had a must at all of those animals brought them back to the yards and put them over a set of scales in the yards and you know we might do that you know once every three months once every six months or once a year if you're lucky but now I get a live weight um for all my animals where they are at the paddock and just to show you what that sort of looks like this is the uh the live weight of some cattle I've just got out the back of my office here and running a small paddock and they're coming across that set of scales every day and I get that live weight update and so you can see I've got animals ranging I've got a bunch of heifers at about 450 kilo and I've got one big old bullock at 840 kilos that struggles to fit up the um in the yard every time we have to right actually run him in um so that's uh that's just a little bit on walk over way and we've actually got some projects over in South America uh in Argentina looking at that technology over there as well and so be keen if anybody's interested to discuss some opportunities for looking at that technology uh in Colombia but the um one of the key technologies that I wanted to focus on were these the on animal sensors and so this is essentially having a smart ear tag uh we call them uh attached to an animal try and hold that up but my camera's probably going to block it so you can see there that's just a simple little um well not simple it's a it's a smart ear tag that contains a tracking device and a radio transmitter that transmits that data back remotely so we can see what the animals are doing out in the paddock so at the moment uh I suspect it's the same in Colombia our producers in Australia we put cattle into a paddock and then we wonder whether they're okay we don't really know for sure but once we have these devices up and working and um then we're able to get that data remotely and actually understand and know for sure what's actually going on with that animal out in the paddock and so there's a whole range of different applications I'm going to skip over that so it's a whole range of different applications of that I'm going to run through that uh shortly and talk a little bit a little bit about what produces that are keen to use that sort of pool in terms of um where things are up to um in the commercialization space there's a whole bunch of um developers out there Phil talked about seris tag they're one of the leading companies out there developing the technology but there's a range of other providers out there so there's a company called movement there we go that's their little tag my camera does not want to play with that so that's one of the other tags um agtech 360 there's another one and there's a bunch more out there and um in the last say maybe six months we've really started to see a lot of these companies companies entering the market and they've got tags out on animals being tested and evaluated and um yeah we're certainly involved in in some of that testing and evaluation just as a bit of an example Phil showed you um seris tags platform before which is um or seris tags data integrated with his platform before here's just another example of a of a similar system from smart paddock and you can log in and see where your animals are and uh what whether they're in the right paddock or not the right paddock but there's a whole bunch of more uh or other applications beyond that um that are of interest to producers and that's the research that i'm involved in looking at taking the data out of this and really turning into something more valuable uh for the for the industry so some of the work that we've done is actually um simply surveying producers to help understand once they had this sort of device if you had a smart tag that gave you the sort of data that you've seen the location and the activity of an animal how could you use it and what we found was that um producers have a range of different applications so they want to use this sort of technology for from simple things like water detecting water related behaviors have they got to the water trough and been able to have a drink or are they being held off or is there no water in the trough and issues like that all the way through to more complicated things like um uh great understanding grazing distribution and Phil talked about that and there's some really big opportunities there in optimizing our landscape utilization what i want to show you though is just some of the work we've done over the years where we've actually worked with producers to put this technology out and let them explore and understand some of the applications that have come to the to the forefront of their minds as they're using um the data basically so this was a property out at long reach which is in western far western Queensland so it's a rangeland almost not quite but uh a real desert uh low rainfall sort of area and in this situation we're tracking two different sheep and you can see one sheep in the bottom right hand corner the green one has got into the water trough which is in the blue circle but the red one just walked straight past and these were weaners so they were a little bit stirred up and not settled down and what ended up happening is that sheep in red actually didn't get into the water for three days because he couldn't find it always being pushed away by other animals and so that's a real production and a welfare concern and so producers are really interested in using it for that purpose uh some of the other things we've looked at is uh stock theft and so stealing of cattle and sheep is a really big issue uh in Australia and in this situation we actually had a a farmer steal his own um animals and so you can see here the green trace this is where the animals are just moving around the paddock quite happily and the red trace is other the records from where the farmer actually got in and started to move those animals to the fence uh to box them up in a in a set of yards and put them on his vehicle and you can actually see him he stole his sheep and took them all the way to his neighbour's place to pretend like his neighbour had stolen his sheep but it showed us that we could actually use this technology to detect this particular issue some other things we've looked at are detecting um plant poisoning issues so we have diseases caused by plant toxicity and we can detect changes in animal behaviour that are associated with those plant toxins and producers once they know that they've got plants in the paddock that are poisoning animals they'll pull those animals out or treat them accordingly and uh and this is the one that uh Phil talked about and I we know that there are some enormous financial benefits to getting this right and this is about understanding landscape utilisation uh so again out in a rangeland environment the uh the red areas are parts of the paddock that were underutilised by the animals and the green areas were the areas that were used quite commonly by the grazing animals and so what uh in this particular case what that actually meant was the farmer decided to put a water trough up one end of the paddock to try and draw animals into that area that was was underutilised and there could be a range of different methods or management strategies that you might use uh to put in once you've got this sort of data so it's sort of there's a whole different a heap of different ways of actually using the information but until you actually have that information you know we don't we don't really understand what we can do with it uh some of the other applications that we're actually researching right now um some of the things that are happening as as I sort of speak um beyond what I just talked about so we do a lot of work in behavioural algorithm development so some of what we're doing is in detecting the carving of cows automatically so you can get an alert say your cows are out there carving um we do a lot of work in basic behaviour modelling as well so just understanding what animals are doing when they're doing it and that informs a lot of the other behavioural algorithm development that we work on um one of the key things that we're working on now is also dystocia detection so not just when that cow's carving or when that sheep is having a lamb but also if there's a problem in that process and so you can be alerted to the fact that there's a difficult carving event actually happening out there uh predation detection so we actually have a lot of wild dogs in Australia that attack and kill calves and lambs and sheep and so being alerted to the fact that there are uh wild dogs attacking your animals and there's some big benefits uh to that to be to allow producers to either go out and directly shoot that animal or maybe target where they put baits out and we do a lot of work now in in disease detection so we're looking at how do we use these um systems to detect some of our endemic diseases so things like bovine ephemeral fever three-day sickness um uh foot rot lameness in sheep and um and worms in sheep as well but also starting to look at how we can use the systems to detect some of our emergency diseases like foot and mouth diseases well in case there's an outbreak in Australia uh bull mating so we're doing some work with um uh Ohio State looking at tracking the activity of bulls and determining when they're actually out there working or if they're broken down and they're no longer serving cows but one of the the main things we're looking at and certainly Phil alluded to this earlier is the idea of integrating sensors so not just looking at um one sensor system to solve the problem but actually bringing data in from multiple places and so having the livestock tracking data linked to the biomass data that Phil's generating and also bringing in the walkable weight data so we know the productivity of these animals as well and we think there's some enormous opportunities around refining things like when to move cattle from one paddock to the next understanding where they're up to in that grazing rotation but there's a whole heap of benefits to come from that data integration so that's something we're looking at um at the moment as well I guess I just want to finish by saying that the smart tags that you see at the moment and the ones that are out there in the industry um we've really only just begun in that space in that technology space and so if you think about it we're in terms of smartphones um you know I walk around with one of the the iPhones at the moment a very clever device but 20 years ago you know the the best we had was a brick phone a really large cumbersome type phone and I think over the next five to 10 to 15 years we'll see a real evolution in these smart tags as they get smaller and more reliable and cheaper to be able to deploy on animals and so we're really just at the beginning of this technology journey in terms of the smart that smart tags that uh that pulls me up thanks folks 319 participants uh joining us uh now uh there are several questions in the in the chat box uh you're breaking up I'm sorry I uh have a connection problem we have some time for questions and there are some in the chat box so how about we uh start with the q&a uh yes Camilo so so we start uh with the q&a Mark and Phil uh if you need to change Salam will ask the questions in Spanish but for you to be able to hear in English you need to change possibly the language in the bottom part where it says interpretation so you can uh you can hear the question in in English again uh Mark and Phil can you hear me in English yes okay uh Andres Fernal has a question in uh mountain lands when we are over 3,800 meters of altitude do satellites work over there oh yeah uh so yes really the only limitation is atmosphere um so uh you know cloud cover uh and the ability to to get through the clouds uh elevation is not a problem okay thank you very much uh for the first speaker uh actor with your technology how can we overcome meteorological problems like rain sand storms fires yeah so there are the satellites that I showed you were optical satellites so they are dependent on on having cloud free imagery the frequency of the satellites we have access to daily satellite data the satellites that we typically use are every five days but we automatically detect clouds and cloud shadows and also atmospheric effects like fire to predict the quality of the data and then inform the user whether there are any issues with the quality of that data before they use it but there are options for for satellite for for satellite radar systems to uh to overcome some of those issues okay or this technology should we count on on good connectivity I suppose I can answer that first mark and you can so for the satellite data um uh short answer is uh is uh connectivity is not really the major issue um we process the data through high performance computing centres and we only send very small amounts of information to the user so you don't need to have very big bandwidth to get access to our information um uh mark we also have the mobile apps the mobile apps work offline and then and then you can collect information offline and then bring that back and go back online to download the data so mark might want to just talk about this of the the IOT sensors yeah so the using the the animal tracking technology connectivity is certainly a challenge uh at the moment there's a range of different solutions if you're happy to have um say low temporal resolution data so uh a frequency of locations say every four to six hours then um saris tag is a direct to satellite connection and so that really overcomes a lot of the problems that we have in terms of getting data out of these difficult landscapes um but yeah in terms of the higher resolution um systems so tags that are working to get a location say every 10 to 15 minutes then you need to set up a local farm network using what's called a laura type system and that requires infrastructure and that does cost and so there's some certainly some challenges around that uh the in all of these companies are solving the problem in in different ways um but there are certainly solutions out there that are working um to some degree in in very communication constrained environments just to add to that um with the saris tag application it's default uh time period is but as mark said four to six hours but it also does actually record baseline activity levels and if if the activity goes outside those baseline activity levels it actually will transmit transmit more frequently and up to uh 15 minutes um if there is actually you know for example if that um tag has suddenly gone from being low activity to traveling at 60 kilometers an hour in a in a truck um uh our system will actually set off an alert and then and then be looking at that tag on a 15 minute basis um uh and sending alerts and emails to the uh to the owner of that animal or that mob of animals uh next question comes from Maria Escobar is it possible for those technologies to recognize the percentage of shadows in uh grazing systems yeah yes uh to that's really part of our remote sensing science so we uh we uh correct for what they call by bi-directional reflectance so um obviously every image has potentially trees and tree shadows uh in it and uh what we do is is attempt to correct that information seasonally we we have an understanding of basically the proportion of shadow for example basically in those canopies that uh that uh we also um estimate the canopy density uh and and then use that to help us estimate um what potential uh pasture is under the canopies uh based on on field observation as well so it requires field data as well as the satellite information to uh to understand the full picture um uh the satellite images do they have good resolution a spectral resolution that that's for your field so yeah so um we look at spectral on spatial resolution so the satellite imagery that I was showing you there was basically 13 different spectral bands so it sees um uh um parts of the spectrum that our eyes can't see uh and and it's 10 meter resolution or 10 to 20 meter resolution but you can still see individual trees uh large trees in that imagery as you increase your spatial resolution so uh google earth for example might be using 30 centimeter resolution satellite data but typically that would only be four bands and it has significant limitations in what you can uh what you can analyze in that data so we compromise by using the central satellites which have 13 different spectral bands of information and 10 meter resolution for which for management is typically uh it's typically good enough but obviously clouds if when clouds are an issue we might need to move to daily data uh which has um higher resolution and higher frequency uh but lower spectral uh resolution okay uh the next question uh a satellite system to monitor animals uh how much is the cost in american dollars and they also request data for uh have a website uh an email address that where we can contact you yeah so um cberlabs.com.au um with seris tags you can go straight to the seris tag website which is seristag.com um and uh they have uh basically introductory packs of of of tags there um you can go and have a look and just confirm the prices i won't confirm it here now but um you can buy uh packs of uh sort of i think 10 10 or 24 tags to start with an applicator to get you going to trial of technology which is what a lot of our clients are doing now is they're buying small groups of you know small boxes of tags putting them on some breeders putting them on some bulls and then starting to learn how to or how to use the technology uh and i'm right now looking at a property um you know 2 000 kilometers away and looking at um you know effectively what animals have actually moved into the waters basically in the last sort of in the last 24 hours so it's it's live data it's three three thousand dollars bill the us for the um uh for that starter kit yep okay as as mark mentioned i think it's really important here where um uh every type of farm has different applications and different priorities and just like mark was showing you know the big brick phone um i think we've all got to um sort of step in there and start you know using the technology and it's only by using it uh and and getting feedback to these sort of providers that the price will come down obviously so there's no point waiting until it's perfect because if you wait till it's perfect you'll actually well firstly it'll never become perfect because you haven't been involved in its development and secondly if you don't if you're not an early adopter it's actually very hard for companies to you know fund the development of that work so it's really dependent on you know the industry such as the cattle industry really sort of you know taking these sorts of new innovations and and trialing them and then working with the other companies to and the universities you know the researchers to develop that technology into into what it needs to be in the future you have some announcements um i would like to remind all participants with us today that from the australian embassy we're going to send an email with a recording all of the presentations and additionally with resources about the educational offer product and services by different australian companies and data contact and we can check that later and i second that motion because later we are going to upload this recording in the fedegan youtube channel and you may consult that to double check and answer your questions um let's see one of the ones that we had in english it had to do with cost the one in english had to do with cost so i think we already answered that one in terms of the satellite so mark talked about the tags in terms of our satellite service so typically the minimum charge at the moment is one thousand australian dollars per year for five boat for five daily satellite imagery but again it's basically it's all about scale and and then so we're looking at a whole range of uh collaborations and partnerships with others that will help us uh drive that price down uh you know by improving the sort of the you know the transaction you know minimizing the transaction costs so um as one example of that um by the middle of next year cibo labs will be providing every red moose red meat producer in australia with a with free access to components of what we're doing um at a farm level so every every month to provide every red meat producer in australia with an estimate of how much partial they have on their property at a property level or a farm level we have another question uh taking into account uh your answer we need to be more useful to use drones algorithms and computers to predict the the amount of dry matter and the nutritional value of uh of different pastures i would say drones would be the most expensive way of doing this um you have to buy the drone you have to buy the batteries you have to buy operators to use those drones and you can only cover very small areas so typically um you know if to cut to cover hundreds of hectares in a single day you would need to have a drone worth you know many tens of thousands of dollars and and then have a an operator to to to use those drones and then to process the data so i would say that the satellite options are many hundreds if in many not many thousands of times cheaper than any any drone application that in terms of extensive pastoral systems okay there is another question that has to do with technology uh the technology that is monitoring different bovines doesn't it mix them up with other animals in the farm uh i can probably just answer from the seris tag perspective so we have the individual animal tags uh we have the uh the device tag as well as the visual tags um so we're we're actually able to look at individual animal tracking over time um maybe your question is perhaps more on the um whether you want to need to tag every animal or only maybe a proportion of those animals and what we're seeing now is you know most people looking at perhaps tagging the very important animals but also then looking at a smaller proportion of their mob to look at what the the overall mob activity mark you might want to um answer some of that as well yeah exactly one last question you're going to elaborate okay no no let's leave it that's good is your system designed to observe behaviors and habits can you estimate uh production and milk production or dairy production or weight production can you detect that also yeah so certainly at the moment we can use that walk over way system to directly detect live weight change in beef animals out in the paddock and that's part of what it's there to do the challenge is that's a quite expensive piece of equipment and so what we're hoping to be able to do um in the medium term medium long term is to actually use the data that feels generating with the pasture biomass and integrate that in with the animal behavior which gives us some indication of how they're how they're grazing using that landscape which gives us a bit of an indication of the nutritional quality of that pasture and then use that to predict the live weight of animals and so um or the live weight change of animals I should say the growth rate and one model we're looking at is to have distributed walk over way platforms in different regions and then use those to ground trip with the bictions but that integration of remote sense vegetation data and the high quality data that feels putting out with animal behavior we believe or at least we hypothesize that we can actually get towards predicting live weight change in terms of beef production at least yeah we're also very excited in these extensive grazing systems around supplementation a lot of our pasture systems depending on seasonal conditions obviously the quality of the pasture varies and and and then what those animals eating are eating in terms of its nutritional value varies so to be able to use this sort of information to then target uh where and when to put in supplementation like lick and those sorts of things could have a fundamental impact on on live weight gain in some of those systems as well Camilo those were the questions that we had there are some that had to do with the value so since we already answered that in terms of cost maybe uh that answer the questions that they had before so uh we should close uh uh should we give them some additional data where you take contact our uh speakers and we can close today uh first perfect sulem uh uh mark uh gave them some links in the chat box but we are going to send the recording of this session to those people registered uh with additional resources contact information educational offered products and services so they can look more in depth uh what is the australian offer and the opportunities there are for uh collaboration uh with a collaboration with australian in this moment so we want to thank you for being here thank you Camilo uh for being with me today and madam ambassador uh eric at thompson uh it's uh my pleasure to see you again after a while and on behalf of fedegan uh we are going to develop these spaces that uh are filled with knowledge and applied with uh with success stories and thank you so much to our speakers today mark phil uh good morning to you and uh this is uh uh we're almost a night time it is six p.m colombia time and we still have more than 260 people we are very pleased by the number and thank you all for being with us throughout uh colombia and the people who join us from different countries uh good evening i'm sula tom and i hope to have you very soon good afternoon