WEBVTT - Farming’s New Frontier: AI Steers Path to Green Future 0:00:00.400 --> 0:00:03.520 This is Dana Perkins and you're listening to Switched on 0:00:04.040 --> 0:00:08.600 the BNF podcast. Pesticides are considered an indispensable part of 0:00:08.640 --> 0:00:13.160 modern agriculture for many farmers. They're used to prevent weeds, insects, 0:00:13.200 --> 0:00:16.639 and fungus from attacking crops and decreasing yields. But the 0:00:16.760 --> 0:00:20.320 use of these chemicals comes at a cost because pesticides 0:00:20.360 --> 0:00:23.560 are a driver of environmental damage and soil pollution, and 0:00:23.600 --> 0:00:26.239 they have been linked to declines in bird and insect 0:00:26.280 --> 0:00:30.680 populations and aquatic biodiversity on today's modern farms. The most 0:00:30.680 --> 0:00:34.600 common way to apply pesticides involves broadcast spraying. This is 0:00:34.600 --> 0:00:38.320 where chemicals are sprayed uniformly over crops, but this method 0:00:38.520 --> 0:00:42.280 is exceptionally inefficient, with ninety five percent of herbicides and 0:00:42.479 --> 0:00:48.239 ninety eight percent of insecticides not actually reaching their intended pest. However, 0:00:48.680 --> 0:00:52.080 new precision delivery technologies could change all of this through 0:00:52.120 --> 0:00:55.520 the use of cameras and artificial intelligence, and it would 0:00:55.640 --> 0:00:59.720 enable green on green sensors to detect and target individual 0:00:59.720 --> 0:01:04.000 weed with pesticides, leaving the other green crops untouched. How 0:01:04.000 --> 0:01:07.080 does this technology work and just how efficient is it? 0:01:07.360 --> 0:01:11.240 And beyond environmental protection, what other benefits does it offer 0:01:11.280 --> 0:01:14.640 to farmers? To find out more, today, I am joined 0:01:14.680 --> 0:01:19.920 by Alexander Littington, an analyst from bnaf's Sustainable Agriculture team. Together, 0:01:20.000 --> 0:01:23.679 we discuss the significance of green on green technology, the 0:01:23.720 --> 0:01:27.560 positive environmental impact it promises, and the potential cost savings 0:01:27.560 --> 0:01:31.120 for farmers. We also discuss the conditions required to use 0:01:31.160 --> 0:01:34.800 these new sensors and whether drones are a viable option, 0:01:35.120 --> 0:01:39.480 And finally, we discuss whether pesticide manufacturers are embracing this 0:01:39.640 --> 0:01:45.479 technological revolution. To access associated BNF research highlighted on today's show, 0:01:45.520 --> 0:01:50.320 including Alex's research note with a terrific title, Precision Pesticide 0:01:50.320 --> 0:01:53.880 Delivery Booming to Spray the Least, you can find it 0:01:54.000 --> 0:01:56.800 on BNF dot com or at BNF go on the 0:01:56.800 --> 0:02:00.040 Bloomberg terminal. If you like this podcast, make sure to 0:01:59.880 --> 0:02:02.920 subscribe or give us a review. But right now, let's 0:02:03.000 --> 0:02:18.360 jump into our conversation with Alex about precision agriculture. Alex, 0:02:18.440 --> 0:02:20.799 thank you very much for joining the show today morning. 0:02:20.800 --> 0:02:22.080 Thank you for having me well. 0:02:22.160 --> 0:02:24.240 Alex is very kindly saying good morning to me, But 0:02:24.280 --> 0:02:27.160 it's actually evening where you are, so I'm located in London, 0:02:27.200 --> 0:02:30.400 you're located in Australia, and we're here to discuss pesticides. 0:02:30.680 --> 0:02:32.400 It would be great if you could give us a 0:02:32.400 --> 0:02:34.880 little bit of context as we head into this regarding 0:02:35.400 --> 0:02:38.919 why we think there might be the opportunity to have 0:02:39.200 --> 0:02:44.200 a breakthrough on being motivated to actually fix the pesticide problem. Because, 0:02:44.280 --> 0:02:47.760 as many people know, in the agriculture space, farming is 0:02:47.800 --> 0:02:52.120 incredibly distributed and so trying to rally behind organized common 0:02:52.200 --> 0:02:56.160 solutions that are centrally managed can be somewhat difficult to do. 0:02:56.400 --> 0:02:58.280 And I think you can give us a little bit 0:02:58.320 --> 0:03:02.040 of context around some of those central sources of motivation 0:03:02.440 --> 0:03:04.400 that are just starting to emerge now. 0:03:05.080 --> 0:03:08.160 So we've got all these new policies emerging, like the 0:03:08.160 --> 0:03:12.519 Global Biodiversity Framework, these big supernational targets to reduce the 0:03:12.560 --> 0:03:14.960 amount of chemical pesticide that we need to apply. And 0:03:15.000 --> 0:03:18.040 this is pretty much the best solution for that, and 0:03:18.040 --> 0:03:20.280 that is to protect nature because obviously all these chemicals 0:03:20.360 --> 0:03:22.760 that leach out and leave the cropping system, they then 0:03:22.960 --> 0:03:27.560 enter the natural environment, They enter organisms, they cause reproductive distress, 0:03:27.639 --> 0:03:30.600 They cause trofic cascade, which is when one species essentially 0:03:30.680 --> 0:03:33.480 falls out of the food web and the whole food 0:03:33.600 --> 0:03:35.080 chain then collapses. 0:03:35.560 --> 0:03:36.960 So it would be great if you could give us 0:03:37.000 --> 0:03:40.640 a little bit of color on the current way we 0:03:40.800 --> 0:03:44.040 are using pesticides in most places in the world so 0:03:44.080 --> 0:03:46.320 that we can understand what it is that actually needs 0:03:46.320 --> 0:03:46.920 to be fixed. 0:03:47.480 --> 0:03:50.680 So agriculture has a trilemma. We need to feed ten 0:03:50.760 --> 0:03:54.280 billion people by twenty fifty on potentially less land as 0:03:54.320 --> 0:03:57.800 it's protected by nature, and to halt the biodiversity crisis 0:03:57.800 --> 0:03:59.960 which we're currently experiencing. As I said, we need to 0:04:00.040 --> 0:04:02.880 feed ten billion people by twenty to fifty. Unfortunately, that 0:04:02.880 --> 0:04:04.640 does mean that we still need to rely on a 0:04:04.680 --> 0:04:08.960 portion of chemical crop inputs as well as implementing integrated 0:04:08.960 --> 0:04:12.600 pest management systems, which are a methodology of reducing the 0:04:12.600 --> 0:04:14.520 amount of chemical pesticide that is used. 0:04:14.880 --> 0:04:16.760 So currently we spray. 0:04:16.600 --> 0:04:21.039 Chemical pesticides at very high volumes, very indiscriminately, and that 0:04:21.080 --> 0:04:24.320 can either be via aerial with a crop duster, which 0:04:24.360 --> 0:04:27.039 is a small plane with a payload that just drops 0:04:27.080 --> 0:04:29.640 the chemical onto the field, or we do it via. 0:04:29.440 --> 0:04:30.919 A big boom spread which is. 0:04:30.880 --> 0:04:33.200 An effect attractor with a tank attached to it with 0:04:33.320 --> 0:04:36.600 two large wings which then deliver that pesticide along the 0:04:36.720 --> 0:04:38.960 rows of crops. Now, what this means is that the 0:04:39.080 --> 0:04:42.200 chemical can actually then either blow away in spray drift 0:04:42.240 --> 0:04:44.960 as the air picks up, or more chemical is released 0:04:45.120 --> 0:04:47.720 then is actually necessary to kill the pest. And that 0:04:47.760 --> 0:04:50.360 pest can either be a weed, it can be a fungus, 0:04:50.480 --> 0:04:54.839 it can be an insect or something even more obscure 0:04:54.880 --> 0:04:57.440 like a nematode which exists in the soil and can 0:04:57.480 --> 0:04:59.960 harm the plant's growth. And eventually what that means is 0:05:00.040 --> 0:05:02.080 the yield that we obtain from the crops. 0:05:02.400 --> 0:05:04.560 So why don't you give me a little bit of 0:05:04.760 --> 0:05:08.840 context around just how bad the problem is in terms 0:05:08.920 --> 0:05:13.080 of the overuse or inefficient use of pesticides, be those 0:05:13.160 --> 0:05:14.600 insecticides or herbicides. 0:05:14.839 --> 0:05:17.839 Absolutely so when we spray these chemicals in these big 0:05:17.880 --> 0:05:21.520 broadcast methods, ninety eight percent of insecticides and ninety five 0:05:21.520 --> 0:05:24.880 percent of herbicides don't even reach their target pests. 0:05:24.920 --> 0:05:26.240 So what that means is. 0:05:26.200 --> 0:05:28.640 It that chemical can then leave the cropping zone and 0:05:28.680 --> 0:05:31.520 the cropping system or the farm and then cause knock 0:05:31.520 --> 0:05:35.040 on consequences to nature and ecology, whether that be disrupting 0:05:35.200 --> 0:05:40.080 breeding cycles of aquatic organisms, killing native plants, affecting the 0:05:40.120 --> 0:05:43.400 behavior of large animals, that neurotoxicity as we call it, 0:05:43.480 --> 0:05:46.000 and even up to the human level. There was traces 0:05:46.040 --> 0:05:49.599 of glyphosate found in human urine at very, very shocking 0:05:49.720 --> 0:05:51.560 levels recently in the United States. 0:05:51.920 --> 0:05:55.520 I think there is a general understanding that for overall 0:05:55.560 --> 0:05:59.200 health of organisms, we want to reduce pesticide use. I mean, 0:05:59.200 --> 0:06:03.480 these are micals designed to kill things, including insects, which 0:06:03.560 --> 0:06:07.560 we do need for biodiversity. So the urgency is so 0:06:07.720 --> 0:06:10.200 well laid out by the fact that ninety eight percent 0:06:10.240 --> 0:06:11.960 I mean, I'm just going to repeat that, because that 0:06:12.080 --> 0:06:16.400 is such a huge amount of overall pesticides not reaching 0:06:16.440 --> 0:06:19.960 their desired location. But Aha, we are here to discuss 0:06:20.160 --> 0:06:23.360 solutions as we always do, and I find this. We're 0:06:23.360 --> 0:06:27.680 going to launch into a conversation around how AI can 0:06:27.960 --> 0:06:30.880 actually address this, which going to be learning a lot 0:06:30.880 --> 0:06:32.840 on this show, because when I think about the natural 0:06:32.880 --> 0:06:36.479 world and I think about plants in particular, the interventions 0:06:36.480 --> 0:06:38.719 that immediately come to mind are not ones that I 0:06:38.760 --> 0:06:42.200 would think involve computers sitting in a lab somewhere and 0:06:42.360 --> 0:06:46.560 a technology fix. So explain to us how AI has 0:06:46.600 --> 0:06:49.440 an application in reducing pesticides. 0:06:49.680 --> 0:06:54.240 Absolutely, so there's a leading technology which is emerging that 0:06:54.320 --> 0:06:56.839 we call as a blanket variable right technology, and it 0:06:56.920 --> 0:06:58.400 kind of does what it says on the tin. It 0:06:58.520 --> 0:07:02.039 varies the right in which we apply these chemical pesticides. Now, 0:07:02.080 --> 0:07:05.880 the pinnacle of that the best performing technology is green 0:07:05.960 --> 0:07:09.640 on green optical spot spray and that essentially is cameras 0:07:09.680 --> 0:07:13.880 that amounted to a conventional broadcast boom spray that then 0:07:14.080 --> 0:07:18.240 uses artificial intelligence algorithms to detect the weeds in the 0:07:18.360 --> 0:07:21.240 row of the crops. As this machine is working down 0:07:21.400 --> 0:07:24.080 the field and it chooses to turn on and off 0:07:24.080 --> 0:07:27.480 the nozzles to only spray the weed that is present. 0:07:27.640 --> 0:07:29.560 So the reason we call it green on green as 0:07:29.560 --> 0:07:33.400 opposed to green on brown was the previous technology was 0:07:33.840 --> 0:07:36.480 a green on brown, meaning that it can sense green 0:07:36.680 --> 0:07:39.840 plants on brown soil. Now we're talking about green on 0:07:40.000 --> 0:07:43.440 green technologies. This means that we can sense green weeds 0:07:43.560 --> 0:07:47.400 amongst green crops. So it's selecting green within green to 0:07:47.480 --> 0:07:49.720 then deliver that precise dose of herbicide. 0:07:49.800 --> 0:07:52.720 So, if I'm understanding this correctly, this is basically a 0:07:53.160 --> 0:07:57.240 very advanced facial recognition technology but for weeds, so it 0:07:57.280 --> 0:07:59.520 can tell the difference between the plant that you want 0:07:59.520 --> 0:08:01.320 to have there and the plant that you don't want 0:08:01.360 --> 0:08:01.840 to have there. 0:08:02.160 --> 0:08:04.360 In essence, yes, okay. 0:08:04.160 --> 0:08:06.240 So this is interesting. I wonder if there'll be any 0:08:06.280 --> 0:08:09.160 mistaken identity moments there. But the point is that this 0:08:09.240 --> 0:08:13.280 seems like a technology solution to detecting the pests that 0:08:13.320 --> 0:08:17.400 are within the desired crops. But why is this considered 0:08:17.480 --> 0:08:21.680 an artificial intelligence application? What about it makes it AI? 0:08:22.320 --> 0:08:25.080 So what they do is they feed a model, a 0:08:25.120 --> 0:08:29.520 computer model, annotated images of weeds thousands of thousands of times, 0:08:29.640 --> 0:08:32.240 and the model begins to learn the shape of the weed, 0:08:32.440 --> 0:08:35.160 the color of the leaf compared to a crop, and 0:08:35.200 --> 0:08:37.160 it can eventually manage to do that at speed and 0:08:37.240 --> 0:08:40.000 make the decision by itself which is the weed and 0:08:40.040 --> 0:08:41.760 which is the crop. Now a lot of this is 0:08:41.880 --> 0:08:45.160 done in house by the big machinery manufacturers, but there's 0:08:45.160 --> 0:08:47.080 also a push to make this open source so that 0:08:47.120 --> 0:08:49.840 it can be used globally free to use, and that's 0:08:49.880 --> 0:08:53.080 by actually a chap working over here in Sydney at 0:08:53.120 --> 0:08:56.400 the Sydney University called the University of Sydney Weed AI. 0:08:56.920 --> 0:09:00.560 So also to simplify this, it enables precision application, so 0:09:00.559 --> 0:09:04.760 it's able to really make the location of the pesticide 0:09:04.760 --> 0:09:08.880 that's being used quite specific to address the specific weed. Now, 0:09:09.120 --> 0:09:12.520 how much of an improvement would this lead to, because 0:09:12.720 --> 0:09:15.840 it's all going into the soil in this general area. 0:09:15.960 --> 0:09:19.319 Is this actually going to dramatically reduce the amount of 0:09:19.320 --> 0:09:22.360 pesticides that are used in a commercial farm? 0:09:22.640 --> 0:09:25.920 Absolutely, the best available technologies are promising an up to 0:09:26.040 --> 0:09:29.079 ninety seven point five percent production in the amount of 0:09:29.120 --> 0:09:30.200 chemical that is used. 0:09:30.440 --> 0:09:32.920 Wow, So that's a huge improvement. And then that begs 0:09:32.960 --> 0:09:37.439 the question does this then have a net economic benefit 0:09:37.480 --> 0:09:41.440 for the farmer? What are the financial savings for the 0:09:41.440 --> 0:09:45.840 farm in terms of reducing pesticides and pesticides essentially a 0:09:45.880 --> 0:09:47.680 really expensive part of farming. 0:09:48.040 --> 0:09:50.600 So yeah, with our modeling that we've done here at 0:09:50.600 --> 0:09:54.120 Bloomberg NIA, we've found that this employing this green on 0:09:54.160 --> 0:09:56.880 green technology can be up to sixty percent cheaper. 0:09:56.640 --> 0:09:58.079 Than current broadcast methods. 0:09:58.160 --> 0:10:01.040 So that takes into account the fuel, all the changes 0:10:01.280 --> 0:10:05.000 in cost of labor, the chemical itself, the herbicide, and 0:10:05.200 --> 0:10:08.720 the cost of the actually implementing and fitting one of 0:10:08.760 --> 0:10:10.680 these new green on green machines. 0:10:11.040 --> 0:10:14.520 So presumably the machines themselves have a pretty substantial capital 0:10:14.559 --> 0:10:17.320 outlaid the beginning, and then the savings that you make 0:10:17.440 --> 0:10:19.840 on reducing your pesticide use is going to help you 0:10:20.000 --> 0:10:22.120 pay for that over time. But let's get into that 0:10:22.200 --> 0:10:24.800 to begin with. How expensive is it? How much cost 0:10:24.880 --> 0:10:27.560 is it going to be for the farmer to start 0:10:27.640 --> 0:10:29.400 using this technology to begin with. 0:10:29.760 --> 0:10:32.400 So there's two models that you can go for here. 0:10:32.520 --> 0:10:35.960 You can either buy a new sprayer which is equipped 0:10:36.000 --> 0:10:39.920 with this technology, or you can be buying a retrofit part, 0:10:40.120 --> 0:10:42.800 which is where a secondary manufacturer will fit this to 0:10:42.880 --> 0:10:46.079 your current existing sprayre Now, most farmers already own a spray, 0:10:46.160 --> 0:10:49.600 they've already outlaid that capsule expenditure. Most of these models 0:10:49.720 --> 0:10:52.160 sit around one hundred and twenty thousand dollars. 0:10:52.559 --> 0:10:55.520 And are these sprayers stationary? Are you capable of moving 0:10:55.559 --> 0:10:58.040 them around? And the real question I'm getting at is 0:10:58.120 --> 0:11:00.240 whether or not you need to have a lot out 0:11:00.240 --> 0:11:02.440 of them on a farm or you'd be able to 0:11:02.480 --> 0:11:05.800 invest in a smaller number and actually move them around. 0:11:06.160 --> 0:11:09.040 Typically a farmer will have one sprayer and that'll be 0:11:09.040 --> 0:11:11.200 the one spread that they use. Now, obviously, in the 0:11:11.240 --> 0:11:13.440 really really large operations, when we get up to the 0:11:13.480 --> 0:11:15.959 thousands upon thousands of hectors. You may need to have 0:11:16.120 --> 0:11:18.480 a couple of machines, but for the most part, one 0:11:18.520 --> 0:11:21.280 machine will definitely take care of all of the land 0:11:21.280 --> 0:11:21.679 that you have. 0:11:22.360 --> 0:11:25.520 So where in the world are these being adopted right now? 0:11:25.640 --> 0:11:30.320 Are there more technology friendly farming communities that we should 0:11:30.320 --> 0:11:32.920 know about somewhere between where you live and I live 0:11:33.000 --> 0:11:34.880 on this beautiful blue planet? 0:11:35.320 --> 0:11:37.000 There certainly are there, certainly are. 0:11:37.160 --> 0:11:39.960 Actually Australia, where we're sitting right now, has had the 0:11:40.040 --> 0:11:44.880 greatest adoption of variable rate technology or precision application. In 0:11:44.960 --> 0:11:48.800 the earlier stages, there was a technology called green on Brown, 0:11:49.000 --> 0:11:50.960 not quite as advanced and can only be used under 0:11:50.960 --> 0:11:54.120 certain applications. It just so happened that those applications fitted 0:11:54.120 --> 0:11:57.880 Australia's agronomy and in geography perfectly, so we had a 0:11:57.880 --> 0:12:01.640 great uptake of the technology and that equates to around 0:12:01.720 --> 0:12:04.760 seventy percent of grain along the eastern parts of Australia 0:12:04.920 --> 0:12:08.360 being sprayed with variable rate technology. Now looking at the 0:12:08.400 --> 0:12:10.880 green on green technology which are starting to become really 0:12:10.920 --> 0:12:13.600 popular here in Australia, but they also will be able 0:12:13.640 --> 0:12:17.160 to provide savings to farmers in the US, in Brazil 0:12:17.360 --> 0:12:19.440 and in the EU. Are the key cropping areas. 0:12:19.679 --> 0:12:22.320 So you mentioned the conditions being right in Australia for 0:12:22.360 --> 0:12:25.040 early adoption of some of these technologies. What are the 0:12:25.080 --> 0:12:27.760 conditions that need to be in place in order for 0:12:27.800 --> 0:12:28.360 this to work. 0:12:28.600 --> 0:12:32.640 Older technologies were only able to spray when they sensed 0:12:32.800 --> 0:12:35.880 a plant and not use AI to detect the difference 0:12:35.880 --> 0:12:38.520 between a plant, a crop and a weed. So that 0:12:38.559 --> 0:12:40.640 means they could only be used when the field was 0:12:40.679 --> 0:12:42.960 being fallowed and that is when the field is empty 0:12:42.960 --> 0:12:45.800 of any crops. So we were only spraying during that 0:12:45.880 --> 0:12:51.240 period between harvest and sewing, again to avoid the cost 0:12:51.440 --> 0:12:55.400 of accidentally spraying your crops. Now with the newer technology, 0:12:55.480 --> 0:12:57.760 we can use it year round because it's able to 0:12:57.840 --> 0:13:02.080 detect the weed in crop. Means whilst the field bears crops, 0:13:02.200 --> 0:13:05.200 we can use this technology to spray throughout the full 0:13:05.280 --> 0:13:06.040 cropping cycle. 0:13:06.679 --> 0:13:09.240 When we're thinking about the sort of crops that this 0:13:09.320 --> 0:13:13.480 would be most used for, whether they're the economically beneficial 0:13:13.520 --> 0:13:16.160 ones because certainly some of them make more money than others, 0:13:16.360 --> 0:13:19.160 or the way that they're harvested does it make a difference. 0:13:19.280 --> 0:13:22.120 I mean, I'm from Napa Valley, which is well known 0:13:22.280 --> 0:13:25.200 as a wine growing region, and there's multiple different ways 0:13:25.240 --> 0:13:28.120 to go about growing grapes, and some of them are 0:13:28.400 --> 0:13:30.760 dry farmed, but most of them are in rows with 0:13:30.760 --> 0:13:33.120 some sort of irrigation. Is this something that would be 0:13:33.200 --> 0:13:35.800 used for the wine industry perhaps, or does it have 0:13:35.960 --> 0:13:39.240 more applicability in let's say corn or even I'm thinking 0:13:39.320 --> 0:13:42.679 about really fast rotation crops near the equator, like bananas. 0:13:43.520 --> 0:13:45.600 So the most gains that we're going to see here 0:13:45.760 --> 0:13:49.040 are typically in row crops. Now, these are big, broad 0:13:49.080 --> 0:13:55.240 acre operations that are growing grains, oil seeds, things like corn, wheat, canola. 0:13:55.600 --> 0:13:57.520 Now that's not to say that it can't be used 0:13:57.520 --> 0:14:01.000 and it isn't used in tree crops. The amount of 0:14:01.000 --> 0:14:02.480 gains that we're going to see are going to be 0:14:02.480 --> 0:14:06.040 best from these larger row crops, big operations. We think 0:14:06.080 --> 0:14:08.200 here in Australia there's some farmers who can go for 0:14:08.240 --> 0:14:10.760 ten kilometers without having to turn their vehicle around. And 0:14:10.800 --> 0:14:13.520 then in terms of frequency of use in the US 0:14:13.600 --> 0:14:17.400 and in THEEU, there's typically one crop per year in 0:14:17.520 --> 0:14:19.960 each field. Now, if we move down close to the equator, 0:14:20.080 --> 0:14:22.280 take Brazil for example, Brazil, you can have up to 0:14:22.480 --> 0:14:25.080 three crops per year being taken out of the field. 0:14:25.160 --> 0:14:27.880 So that really equates to a lot of savings on 0:14:27.960 --> 0:14:29.520 the amount of pesticide that we're spraying. 0:14:29.960 --> 0:14:32.760 So you've painted a picture in my mind of fairly 0:14:32.960 --> 0:14:36.840 large farms monocropping, and you know, the things that are 0:14:36.880 --> 0:14:40.240 actually reaching most people on their dinner plate. And there 0:14:40.280 --> 0:14:43.640 have been concerns not just regarding pesticides, but actually the 0:14:43.680 --> 0:14:47.040 vulnerabilities of monocropping and also what it does to the 0:14:47.080 --> 0:14:50.600 soil to constantly be rotating the same crops on the 0:14:50.640 --> 0:14:53.600 same space. Is there a benefit to soil health when 0:14:53.600 --> 0:14:56.120 it comes to reducing the amount of pesticides And is 0:14:56.120 --> 0:14:58.040 this going to make that plot of land that the 0:14:58.080 --> 0:15:00.400 farmer is working on something they're going to be able 0:15:00.440 --> 0:15:03.120 to be on for longer and maybe even reduce their 0:15:03.280 --> 0:15:04.480 need for fertilizers. 0:15:04.840 --> 0:15:05.280 Certainly. 0:15:05.360 --> 0:15:08.360 So, a lot of these chemicals reduce the health of 0:15:08.400 --> 0:15:12.600 the soils, not only in the microorganisms, but also those 0:15:12.840 --> 0:15:17.360 larger organisms that are existing, the beneficial insects and invertebrates. 0:15:17.640 --> 0:15:20.520 So by reducing the amount of damage that we're causing 0:15:20.680 --> 0:15:24.440 to the soil by overspraying pesticides, we actually stop the 0:15:24.520 --> 0:15:28.440 vicious cycle, which is damage soil less beneficial insects and 0:15:28.480 --> 0:15:30.520 therefore more chemical needing to be sprayed. 0:15:32.160 --> 0:15:36.240 The pesticide business is a big business. By reducing the 0:15:36.240 --> 0:15:38.840 amount that are actually being used, there could potentially be 0:15:39.120 --> 0:15:41.280 a lot of money at stake. What is the current 0:15:41.400 --> 0:15:44.160 market size for this and how much are they doing? 0:15:44.360 --> 0:15:47.240 I guess annually in terms of how many pesticides are 0:15:47.280 --> 0:15:48.960 being sold around the world. 0:15:49.280 --> 0:15:53.160 So the global pesticide industry or crop protection industries is 0:15:53.200 --> 0:15:56.000 also known, is around forty three billion dollars a year. 0:15:56.120 --> 0:15:58.920 Now that's completely under threat at the moment, both due 0:15:58.960 --> 0:16:01.680 to legislative pressures but also due to the farmer being 0:16:01.760 --> 0:16:04.400 very squeezed. The farmer doesn't receive very much of the 0:16:04.400 --> 0:16:08.200 share of the dollar, so wherever they can reasonably cut costs, 0:16:08.320 --> 0:16:11.560 they will. Now this technology is allowing them to reduce 0:16:11.760 --> 0:16:14.800 both their cost base but also to improve the productivity 0:16:14.840 --> 0:16:15.840 of their farm. 0:16:15.960 --> 0:16:19.840 How are the pesticide companies viewing this new technology that's 0:16:19.880 --> 0:16:22.240 emerging and is it something that they're interested in and 0:16:22.280 --> 0:16:25.960 potentially developing a house or even acquiring, or is it 0:16:26.000 --> 0:16:29.800 something that they're threatened by and trying to figure out 0:16:29.880 --> 0:16:33.040 how long they can avoid it becoming mainstream. 0:16:33.320 --> 0:16:36.840 The crop protection industry has been able to move very agile, 0:16:36.880 --> 0:16:42.040 and they've learned from previous legislative knocks such as in 0:16:42.120 --> 0:16:45.360 the EU the amount of pesticides that are being restricted 0:16:45.360 --> 0:16:47.400 for use on a yearly basis. So they've really grabbed 0:16:47.440 --> 0:16:49.440 the ball by the horns and they've jumped in at 0:16:49.440 --> 0:16:52.360 the deep end. For instance, there's a startup called the 0:16:52.440 --> 0:16:56.440 Eco Robotics ARA, which is a precision delivery unit which 0:16:56.480 --> 0:16:59.680 promises to reduce crop inputs from herbicide insect side and 0:16:59.680 --> 0:17:02.760 funger side as well as fertilizer. Now that's had investment 0:17:02.960 --> 0:17:07.280 from notable investors, both the ASF, the German crop protection company, 0:17:07.320 --> 0:17:10.400 but also Yara, the global fertilizer company. 0:17:10.880 --> 0:17:13.600 So when you're talking about this AI technology being used 0:17:13.640 --> 0:17:18.400 to spot those pesky pests, you specifically referenced it being 0:17:18.560 --> 0:17:22.880 put onto a boom and kind of coming across the 0:17:23.080 --> 0:17:27.040 entire field. But I'm thinking now of drones that are 0:17:27.080 --> 0:17:30.679 not actually connected to the ground flying over. That is 0:17:30.720 --> 0:17:34.159 something that has been i think excitedly talked about. And 0:17:34.200 --> 0:17:36.439 the question for you, is someone who's looked at this 0:17:36.560 --> 0:17:40.040 in more detail, is whether or not you think that 0:17:40.040 --> 0:17:42.200 that's got legs No pun intended. 0:17:42.720 --> 0:17:48.000 So there absolutely is an application for drones in agriculture. However, 0:17:48.160 --> 0:17:52.440 the spraying of large sways of cropping land is probably 0:17:52.480 --> 0:17:54.879 not it. The payloads are just too small at the 0:17:54.920 --> 0:17:58.440 moment and we haven't really seen advances in swarming technology, 0:17:58.440 --> 0:18:01.399 and that's when we deploy multiple drones. So I think 0:18:01.480 --> 0:18:05.080 the largest drone which is currently available is the Guardian 0:18:05.119 --> 0:18:09.040 Agriculture sc one, which has around seventy to eighty liters 0:18:09.280 --> 0:18:12.600 of payload capacity. It has also been invested by big 0:18:12.720 --> 0:18:15.800 names in the industry like leaps by Buyer. Now, the 0:18:15.920 --> 0:18:19.720 case for drones is that we can apply chemical to 0:18:19.960 --> 0:18:22.400 hard to reach areas, which will mean that the use 0:18:22.400 --> 0:18:25.399 of the chemical is more efficient, so such as the 0:18:25.600 --> 0:18:28.720 tops of tree crops, which previously we haven't been able 0:18:28.760 --> 0:18:32.280 to apply to very efficiently we've therefore lost yield. It 0:18:32.320 --> 0:18:35.520 also can be used on areas such as boundaries where 0:18:35.720 --> 0:18:36.720 the spray is. 0:18:36.640 --> 0:18:38.280 Not able to get to as easily. 0:18:38.640 --> 0:18:43.480 Alternatively, and areas of sprayer can't enter, like greenhouse agriculture 0:18:43.520 --> 0:18:45.879 or glasshouse agriculture. So there certainly is a case for 0:18:45.920 --> 0:18:48.720 spray drones. The technology is just not there yet at 0:18:48.760 --> 0:18:53.359 the moment to spray big swathes of agricultural arable land. 0:18:53.760 --> 0:18:57.120 Another application of the use of drones will be in 0:18:57.240 --> 0:19:00.520 mapping and surveillance of fields. So that is something that 0:19:00.640 --> 0:19:03.280 currently takes a lot of time and effort and also 0:19:03.280 --> 0:19:05.520 a lot of money to get a satellite map done 0:19:05.680 --> 0:19:08.959 of the characteristics of your farm land, which are very 0:19:09.000 --> 0:19:12.240 useful in applying inputs. So that may be the yields, 0:19:12.320 --> 0:19:15.200 that may be the presence of pests, that may be 0:19:15.480 --> 0:19:16.680 the availability of. 0:19:16.760 --> 0:19:18.199 Water for your cropping. 0:19:18.240 --> 0:19:22.320 So this will enable us to work smarter using imagery 0:19:22.440 --> 0:19:25.400 mapping software that then can be loaded up to farm 0:19:25.400 --> 0:19:29.159 management software to deploy chemicals smarter. So the case for 0:19:29.240 --> 0:19:34.280 drones is either as surveillance and mapping imagery technology or 0:19:34.400 --> 0:19:37.520 as in small payloads in areas that we can't get 0:19:37.520 --> 0:19:38.280 a boom spread to. 0:19:38.760 --> 0:19:41.560 So the drone technology is actually a compliment to the 0:19:41.560 --> 0:19:44.760 boomsprayer as opposed to an alternative to it, at least 0:19:44.760 --> 0:19:47.040 in the picture that you've painted of the future, where 0:19:47.080 --> 0:19:49.800 we're looking at the field and being strategic and then 0:19:49.800 --> 0:19:53.280 the boom sprayer has that ability to actually do the 0:19:53.440 --> 0:19:56.840 end application in that quite precision way. This seems like 0:19:56.880 --> 0:19:59.399 a great solution. You've got me very excited about what 0:19:59.440 --> 0:20:02.360 this technology could accomplish. So let's pivot a little bit 0:20:02.480 --> 0:20:06.360 to what parts of the pest world that it could 0:20:06.400 --> 0:20:09.120 actually be applied to. A lot of our conversation thus 0:20:09.160 --> 0:20:14.439 far has really focused on weeds and the plant based 0:20:14.560 --> 0:20:17.240 problems that we have there, but invariably there are also 0:20:17.560 --> 0:20:23.480 fungus and insects. Is that the next frontier for this technology? 0:20:23.560 --> 0:20:26.560 And I guess, without asking you to predict the future, 0:20:26.760 --> 0:20:30.040 how soon do you think it'll be applicable to a 0:20:30.200 --> 0:20:32.200 wider range of pests? 0:20:32.600 --> 0:20:36.639 Absolutely so, we already have companies offering drone surveillance to 0:20:36.840 --> 0:20:39.840 be able to detect other types of pests, such as 0:20:40.400 --> 0:20:44.080 Xavio for instance. However, what we haven't yet seen is 0:20:44.440 --> 0:20:47.960 a machine that is actually capable of seeing and detecting 0:20:48.040 --> 0:20:51.160 the other types of pests in real time and then 0:20:51.280 --> 0:20:55.320 therefore delivering the precise dose of required pesticide, whether it 0:20:55.400 --> 0:20:58.280 be a herbicide, insecticide, or fungicide. Now we've put a 0:20:58.400 --> 0:21:01.200 rough estimate there together that we think by at least 0:21:01.240 --> 0:21:04.320 twenty thirty we'll have this detection available for fungicide, so 0:21:04.359 --> 0:21:08.080 that disease will be detectable by a machine in real time, 0:21:08.280 --> 0:21:11.280 and then for insects, we think that should come land 0:21:11.440 --> 0:21:14.920 somewhere between now and twenty fifty. The problem with insects 0:21:15.040 --> 0:21:18.400 is that they are mobile, they're fast, and they're cryptic. 0:21:18.680 --> 0:21:21.040 That means that they can hide, so actually spotting them 0:21:21.040 --> 0:21:23.760 in a crop is a particularly difficult endeavor. 0:21:24.160 --> 0:21:26.919