Like the data transparency is clearly not solved. Well, I'll tell anybody on this podcast today, because I'm sick of fighting about this. The government of Brazil, In-Man, has a map, which allows you to choose weather station networks to view rains over the past three hours to the past three months. You can look at it and you can find it and it's free.
It is free. They are giving you the data. You should be able to find that. You don't need some guy with a broker report or some guy with a written report or some origin to tell you what's happening when you can look at it yourself.
And I believe that everybody should be having a vested interest in educating themselves on the weather and the open source data side of it. You shouldn't need me. I'd like to tell you what happened. We have some exciting new changes coming to Map of 40 in 2026 to support the rapid growth we're experiencing around the world.
After seeing the impact this podcast has had on businesses and professionals in the coffee industry who have access to it in English, we've decided that we'll be working hard to make it available in other languages that encompass our amazing and diverse industry. To fund that growth for the very first time in a very long time, we'll be inviting advertisers onto the global podcast and the Map of Forward Middle East podcast. So if you're a small to medium sized company and looking for B2B advertising opportunities and you believe that our values are aligned, please send us an email at support at mapofour.org or DM us on social media and we can start a conversation today. Check the show notes for links.
Welcome to the Daily Coffee Pro by Map of Forward Friends. I'm your host Lee Safar and this is episode four of a five-part series with Greg Weatherstrategist Greg from Sukafina. We are talking about the role that weather is playing in the coffee market. And in this episode, we're talking about data, the need for open source data in the coffee industry.
First of all, Greg, what does open source mean? Yeah, it means that anybody can get their hands on it and not have to spend a lot of money. You need a computer, preferably it's free. You have a computer, now I guess you use AI to code for you and they figure out and they get the data and you're able to see the data and come to your own conclusions.
And for a long time on the weather side, that has been a source of frustration for me. Dealing with third party vendors, dealing with people who are taking data and repackaging it and making it easy to use. And increasingly what we're starting to see is people are aware that people will scrape this data, like people on weather stations are aware that this data will be scraped. It's been really refreshing to see the way the Europeans have handled it, the ECMWF, which is obviously the proprietor of the best weather model we have out there.
But also they've become really, really a lot more open source with their data over the past few years. And that's been refreshing to see that, you know, from a weather modeling perspective, they're making more available to people. And people like me and others to be able to get data and get an idea of what things are being said. Where we're running into trouble, in my opinion, is observational data on the weather.
I think that has been an area while the ECMWF has produced a five day lagged data, big data set. Rural data? Yeah, it's basically a snapshot of the world at any given time. It's called the RA5, and it's kind of become like the gold standard of what we use in trying to figure out this data, but it's not ground station data in all these places.
And I think what I'd really like to see is investment in ground station observational data, quality controlling that data, making that available to people. So people can understand what's happening at any given moment in time. I think part of what drove you to speak to me and how you were able to kind of get to reaching out to me is because I had a very, not so, I guess, censored view about the way that data is being presented to be. In particular, when it came to, you know, looking at how rains have fallen in Brazil, I think that would be kind of how I would say is what really kind of drove our initial conversation.
I haven't been shy about the fact that I did not like the way people presented data in Brazil this year. What did you like about it? I don't like when weather stations conveniently go offline at times that people are looking at them. I don't like when people are seeking and cherry picking selective weather stations that may be not working or maybe working for a short period of time and saying it only rained eight millimeters in the first half of October in Brazil.
When every frickin' person knows that's not true. But if you're selling a bullish story to people and other people aren't doing the work to look or they don't know where to find it, that sounds plausible. It's not gonna be manipulation. That sounds as plausible as anything I'm saying.
So I wanna fight with people about what's going to happen. He comes to Italian. I wanna have an argument in the discussion with people about what the view is two weeks, three weeks, a month from now. It is a waste of my time, your time, and everybody else's time fighting over did it rain in Brazil.
That's frickin' stupid. That is a stupid argument to have. Especially because it's an objective fact that we can figure out, did it rain or didn't it rain? Like it's not really- Look outside your window.
Look outside your window. Did it rain today? Yes or no? You can tell me that.
Exactly. But some dude can just take his weather station offline, put out a big report and say, oh, it didn't rain today. Meanwhile, he's walking in tunnels when he goes outside. And if the data's not transparent, his weight of his words can, or her words can influence things.
We have no shortage of people who have agendas in this industry. I have certainly been accused of having an agenda myself. But I can tell you one thing that I do not do, is I do not misrepresent what the temperature was or the rain that has fallen. We have the ability, we have enough stations to be able to make that assessment on our own without having to rely on a broker board or a bank report or something like that that tells you something didn't happen when we know it did.
And I think that's really what got me bothered in September and October this year, was like, I went to LinkedIn. I know where else to go. I went to LinkedIn and I'm like, here's the data. You know what?
You're not getting the data. Here you go. This is what happened. These are the rings that fell.
This is what happened in the past 14 days. This is what happened in the past 30 days. And if you're an end user or you're someone who's taking delivery of coffee or if you're a producer and you don't know where to look, I'm the guy who has to tell you. I'm the meteorologist.
And then I found you. I'm the weather strategist, right? So here, there you go. I guess I'll be the open source data guy for everybody.
Or anybody who wants to do it. I still get somebody sending me a link to the message. Like, what happened? What are you seeing it?
And I'm like, all right. So this clearly is not a solved problem. Like the data transparency is clearly not solved. Well, I'll tell anybody on this podcast today, because I'm sick of fighting about this.
The government of Brazil, InMet, has a mat, which allows you to choose weather station networks to view rains over the past three hours to the past three months. You can look at it and you can find it and it's free. It is free. They are giving you the data.
You should be able to find that. You don't need some guy with a broken report or some guy with a written report or some origin to tell you what's happening when you can look at it yourself. And I believe that everybody should be having a vested interest in educating themselves on the weather and the open source data side of it. You shouldn't need me on LinkedIn to tell you what happened.
Now, not everybody's going to do that work. I can lead a horse to water, but I can't make a horse drink. But the transparency of the observational data, taking care of our weather networks, making them easily available and accessible. So people know what is happening.
To me, I don't see the upside in that. I don't see the upside of hiding that, unless you haven't done. Let's fight about what's going to happen. Yeah, the speculative side of it, that's an arena that people joust as much as they want.
But the stuff that's actually happened, well, exactly in that side of it, but the stuff that's already happened, if you're trying to keep that information from people or lie about that information, one, why? And two, it sounds like that's for market manipulation. It sounds like you're trying to trick people and fall people. Yeah, I don't like to operate in that space about trick people about what's happened.
And not all of it's nefarious. Sometimes weather stations go offline. Sometimes things don't work. We know that.
But, you know, what I don't like is when you cherry pick data to tell a story. Like, just tell everybody, just tell everybody. Well, that's just shit science. Yeah, show what's going on.
You know, you have enough. You have enough. And I also think that we are starting to see private investment in weather stations to try to fill in these gaps where governments can't do it. I welcome that.
And I understand that everybody's got an ROI that they need to hit, but like, we should be able to know if it rain. Like, that's... Yeah, that's basic. That's basic shit.
It is. I have a strange question for you that I didn't even think about until you started talking about this. Do you know surely you know what a farmer's almanac is? Of course.
For those who don't know what a farmer's almanac is, could you explain it to people? Yeah, I mean, it depends on where you're looking. There's the old farmer's almanac, which is like a little bit of pseudoscience, right? Which is what's going to happen the next three to six months, which is just based on who the hell knows where they come up with it.
But, you know, it's meant to be a useful... A prediction mechanism, right? A prediction mechanism. Sort of open-sourced data that everybody can sort of can get access to.
The methodology I won't subscribe to, but at the very least, you know, they're sticking something out there. I think that that side of the equation, like the farmer's almanac, is like, it's fun. It's fun to look at. We are in 2025.
We are now able to view observational data. Actual data, right. We can look at it. Let's expand that.
Let's see. Hey, how are the rains in the past 90 days in Honduras? Like, how are the rains in the past 60 days of Brazil? What's the temperature been?
You should be able to find that without having to repurpose and repackage it and sell it to thousands of dollars for people. I just have a fundamental disagreement. Let's have a... Again, let's have the debate about what's going to happen.
Let's not have the debate about whether it rains. Come on. That's a waste of my time. It really, it's a waste of everybody's time.
But the reason I ask you about the farmer's almanac is because I've spoken to so many farmers who said, you know, you used to be able to set your clock by the farmer's almanac. You used to be able to know exactly what day. And farmers almanac, folks, would be sort of like a calendar that would tell you, you can expect it to rain this many days of this month, in the year, the rains are going to turn up this time, then there's going to be a dry season, then it's going to rain again. Whatever it is, the farmer's almanac was given...
It was sort of created by the people who would understand the weather by the farmers in those regions, because they knew those lands, they knew the weather, the weather was consistent, and they published it, and they passed it around. That was like the open source data of that time. Is the reason that those farmers almanac, even though they still publish them, is the reason that they're getting less correct because of the changing weather? I think also there's one thing to remember, Lee, is that the more information you find out, the more you can test the theory of what you're reading.
Okay. So if you're in a region where rainy seasons normally begin within a one to two week period, or often than not, that's what's going to happen. If you're in an area where that two weeks becomes three, or if you're in an area where you have more information of what's going on around you and things of that nature, maybe you find out they're not the most accurate things before, right? So I think that as people become more aware of what's going on around them, I think you strip away the, oh, this is just the, what's going to happen, the gold standard of what's going to happen, it's perfect.
Hell, when I was growing up, it was the same thing. People were like, oh, it's going to be cold winter. I'm in the northeast of the U.S. We all give a shit about snow, it's like snow and cold.
When it's going to be a cold winter, that's what we care about. That's what a lot of people care about up here. But I think, ultimately, there's no silver bullet in what I do. There's no silver bullet.
You can set your clock to this is what's going to happen. There's variability in each year. I think we just become more aware of the availability. And yes, there's been certainly more volatility.
I think I don't think anybody would argue that. Maybe somewhat, but I'm not one of them. But I think that the more information you get, the more you can look back in prior history, the more informed you become, the more you can test the theories of what you're being sold. And maybe the relationships don't hold up as well.
So we're going to open source data, right? Let's say that all of this data is available to people. It sounds like that as a weather strategist, your job is to look at the data and to be able to interpret that data. And is that where the science comes into it, right?
That's where the, we can all look at the same thing. And if we don't understand what we're saying, we're not going to know how to apply it. Am I understanding? I think that applies to more than just the weather.
It applies to a lot of... We were having a conversation about politics. A lot of things that are going on right now, totally. It's all a conspiracy to do.
You don't know what the hell's going on. Even the weather now. Yeah, I think, you know, what people don't, I think maybe understand about what I do and what meteorologists do is that we have to be comfortable with the unknown. And we have to be comfortable with the fact that we are in a system that we do not have appropriate predictability on.
There's an old adage. I can get it still three to four times a year, although I get it less now than I used to, which is, whether it's the only thing you can be wrong 50% of the time is still on a job. I can't be wrong 50% of the time. No, I can't.
I won't survive in this industry if I'm wrong 50% of the time. But what I can do is help people understand about the way I see things evolving. If the wrong answer, if I get the wrong answer, but I know that things are going to happen beforehand and people can prepare for that, you know, maybe I'll get my seasonal forecast wrong by 100 millimeters of rain, but I was able to guide you along the way of like, okay, we have a dry period here, we have a wet period here. On the aggregate, this is going to be okay.
This is not going to be okay. Then I'm doing my job. And I think that where my job comes in is not just interpreting the data and using the 15 years now I've been doing this to understand how I see the crop evolving, working with agronomists to hear what they're seeing, kind of understanding the way that work streams are changing the way that we have to view coffee, all of those things. It's also the art of forecasting, the art of understanding, look, I recognize this pattern.
I've seen this pattern before, but maybe I tweak it a little bit for a little bit of a climate change that I'm observing that's going on with it. It may not be the most scientific thing in the world. There's no weather model that's 100% right. A lot of my job is figuring out where the hell they're wrong.
A lot of my job is saying, okay, this is what the market is seeing. This is the weather model that the market is seeing. That park doesn't make its own sense 10 days from now. Maybe that's going to come in this way.
And maybe they're not going to see it now, but they're going to see it five days from now. And maybe that's going to be how the market reacts to it. So I think open source data does a job of letting you know what has happened. It's not necessarily going to mean that you're going to become the world's greatest forecaster and everybody's going to figure out what's going to happen.
There's still an inherent skill to what I do. There's still an inherent understanding of things that I've learned through years of experience, years of schooling, all those sorts of things where I can apply what I know in a way that makes commercial sense and makes agronautical sense. So I'm taking complex things, making them very easy for people to digest, explaining how I see it evolving. I think this is what is going to happen, which means this to perception, which means this to the market price.
That is kind of where I bridge the gap between trading and whether I'm a trader myself, like I'm a future trader, I've traded five years, four years of futures at my prior shop. I do a little less of that now, but I still have a very big role to play in how we decide to set our position. I have a very big role to play in how we figure out when we're going to buy coffee, what qualities we're going to buy how much, off of what I think the weather is going to do. Open source data will just at least help us monitor things.
So if we can see if our belief is rooted in fact or not, like you're getting new data every single day that either is helping or hurting your hypothesis and you have to change your view if things observationally telling you that you're wrong. So I think humility to know when you're wrong, the humility, that's where open source data helps. The humility to know where you're wrong and you need to pivot or giving you something that you can look at objectively, which allows you to inform your future. I think that is how I see open source data really being important to our industry because farmers should know.
Farmers should have a really good idea of how much is going to rain. My yield potential, maybe at the beginning of the year, I could make 500,000 bags or I could make 5,000 bags or 3,000 bags. If the weather has done this, maybe that goes from okay, I can make 4800 bags to 4,100 bags. The weather goes forward.
Okay, maybe it's 4,500 to 4,200 and you just narrow the amount of uncertainty that you get forward with time with observed data. And if you have that observed data, you have a better way of figuring out what you think is going to happen that informs your marketing strategy if you're a farmer, then informs how much you think you're going to be able to make, how much you need to invest next year, what you think your production is going to be. And it helps so much for informing the business decisions that make this whole thing run. So I'm someone that believes that observational data, we all should have access to it.
We should all know what's going on. And it's up to you to get that observational data. I can't do that for you, but I do think that it's not going to solve necessarily everything in terms of what's going to happen. How far ahead can you typically look at something and say, I think this is going to happen then?
It depends on what I'm looking at. Right. I can have a pretty good idea of an El Niño or a La Niña. And what that is likely going to mean, six to eight, nine months in advance.
Well, I can have a pretty good idea, if I see a La Niña coming in two months, I have a pretty good idea of five to six months from now, what global things that means, that will mean a wet column the other side of it, or a wet unit. I saw, for example, and I flagged that, the minute we were in this La Niña this year, and I saw that we were going to it in the spring, I said we're going to have problems with our hearts. That doesn't mean I'm going to know it's 30 or at five degrees at this point. But you know the direction.
I can at least have an idea or I have a statistical odds that this is much higher probability of happening this year than last year, or this year versus relative to normal, which is part of the job. If I'm trying to figure out how much it's going to rain in a specific timeframe, in a specific place, then you're talking like, you know, two, a week, two weeks, days, you know, you start getting into that. But for us, our frost is going to happen. Most of the industry didn't think the 2021 frost was going to happen five days out.
I'm like, I don't like the way this looks. This looks like, if it's going to go wrong, it's going to go wrong. It's going to go wrong. Really wrong.
And that's what I told Dave, and Dave Aaron is our trader and everybody, and Dave goes, you've never sounded like this. I've never seen something that's not in much, since I've been forecasting, not like this. I've never seen top-end potential that's going to go wrong. So it depends what you're forecasting, when you're forecasting, what you're trying to understand and figure out.
And I think that predictability in terms of broad strokes is, yeah, like I said, I can have a pretty good idea if it's going to be a wet harvest. If I know that there's a lunging, you're going to be there. But if I'm trying to figure out the specific temperature, the specific, you're talking about a much shorter timeframe. Wow, fascinating, fascinating.
We folks have one episode left, and it's going to be the last episode of 2025. So join us in the final episode of this series. We are asking the question, what are the biggest challenges for coffee in addressing the impact whether it's having on-coffee production? So join us for the final episode of 2025.
Couple of us said that. Peace on peanut butter, I have an amazing rest of your day. This episode is part of a five-part series produced by Mapper Forward. Check the full playlist and guest details in the show notes as well as how you can support our work in the coffee industry.
Become a premium YouTube subscriber, explore our Patreon and any brands we've partnered with to bring you this podcast. Please don't forget to like, share and subscribe. It really does help us grow our audience.