From GPS points to satellite data: The future of traceability with Nicole Linares

This regulation requires companies that sell certain commodities in the European Union to demonstrate that those products were not produced on land that was recently deforestated, for example, when we were on the field. And when enemies start tracing the boundaries of their farms on the tablet, farmers are often surprised to see their land represented on a satellite image. First, what I actually do is process the data using Python and goodies, which is extremely useful because it allows you to organize and visualize all the farm layers in a very practical way. Also, something that has caught my attention lately is that there are some advances in satellite mapping using machine learning models developed by Google to analyze large volumes of Earth observation data. Today on Survey and Beyond, we are joined by Nicole Linares, research analyst at letright. With extensive experience in data collection and geospatial analysis, Nicole has worked on projects across Africa and Latin America, helping organizations like Rainforest alliance meet sustainability and compliance goals through offline geospatial data collection. So she's here to show us how geospatial data collection can empower farmers, ensure regulatory compliance and unlock new opportunities for market access. Welcome, Nicole, to Survey and Beyond. Hi Marta, thank you for having me. Yeah, it's a pleasure. So let's kick off our conversation. Could you start by sharing a bit more about yourself and your journey and how you got involved in this field? Yeah, of course. My background is in research and data collection for development projects, particularly in rural contexts in Latin America. I currently work at letter 8 with this research data firm that supports organizations working in areas like agriculture, supply chain and environmental sustainability. Awesome. And what does your work at Letright mean exactly? So what do you do typically in your work particularly related to data collection, which is our main theme for this podcast at Letrate, I work mostly on the implementation side of research projects. So that includes designing surveys, setting up data collections systems, supporting field teams, and monitoring data quality during field work. Many of the projects we work on involve collecting both survey data and geospatial data. For example, this case we have with Rainforest alliance, it was all about mapping farm boundaries and understanding land use patterns. So I will say that a big part of my work is making sure that the tools and workflows we design actually work in real field conditions. Okay, great. And I'm glad you already talked about your project related to the Rainforest alliance, because that's going to be our focus today. So I'm going to briefly very high level describe it so and let me know if I am mistaken at any point, but you have been working in a project to gather data for the Rainforest alliance that is supporting cocoa and coffee farmers in Peru meeting compliance with the EU deforestation regulation. And because in this sentence I mentioned so many terminologies, I feel like it's a good idea to just define some concepts for our listeners. Would you be able to explain the Rainforest Alliance's work? What is the EU deforestation regulation and how they intersect? Yeah, about the Rainforest Alliance. They work with farmers, communities, companies and governments to promote more sustainable agricultural production in the context of coffee and cocoa. They support farmers in advance practices that protect forests while also improving productivity and livelihoods. More recently, there has been a lot of attention on supply chain transparency, especially with the introduction of the EU deforestation regulation. This regulation requires companies that sell certain commodities in the European Union to demonstrate that those products were not produced on land that was recently deforestated. So there's now a much stronger need for reliable data about where farmers are located and how land is being used. And when we are talking about being compliant with certain regulations and certifications. At least me personally, as a buyer in the EU market, usually I think about how we go to the supermarket and in every product or in every package, you see a physical seal or you see a logo to help you guide in your decisions as a buyer. But I think this is not the case for eudr. Like it's not an optional certification in any way. It's just something that is really mandatory. It's a requirement for something to be sold in the EU market. Right. So it's really important and impactful for farmers. Would you be able to share what do farmers need to do to comply with this regulation? Yeah, in practice, it means that supply chains need much stronger traceability systems than before. Companies must be able to demonstrate exactly where commodities are coming from and verify that they were not produced on land that was being deforestated after the regulatory cutoff date. To do this, they need reliable information about the location of each farm and the precise boundary of the production areas. For farmers, this often means that their farms need to be properly mapped and registered within traceability systems. Accurate geospatial data, such as GPS coordinates or polygon maps, becomes essential to demonstrate this compliance. So in practice, modern farms accurately and maintaining good quality location data is becoming a key step for producers and farmers in general who want to keep assessing European markets. That makes a lot of sense. And I feel that's exactly where you come in. Right. Collecting that GPS data that you were talking about doing those boundary mappings. Could you briefly describe what this data collection process entails? Of course. In this project Our team visit farmers and collect several types of information. First, we conduct surveys to understand farming practices and socioeconomic conditions. But at the same time, we map the boundaries of the farm using GPA's based tools. Instead of just recalling a single GPS point, enumerators walk along the perimeter of the farm to capture the full boundary of the plot. And in cases where the land is too large to walk, they map the boundary directly using satellite imagery, tracing the plot with the farmer's guidance. I think this allows us to generate much more accurate representation of the farm and where the production is taking place. And out of curiosity, so that you know that no deforestation has happened, like over time, is this something that you need to collect longitudinally? So from time to time, just to make sure I know no deforestation has happened? Yes, it usually is collected annually, and that allows field teams to be prepared for this data collection, but also to farmers and organizations to be aware that this time of the year are going to be collected data of the polygons. And in that sense, if it's annually, you can have a much more traceability system through all the years. Absolutely. And you were mentioning that they were collecting several geopoints so that they create a polygon. If the land is too large, they would use satellite imagery. So like a single GPS point is not good enough now. Right. You really need to map the whole land of a farmer so that you can accurately track the deforestation process, if there is any. Yeah, exactly. For that reason, I think a point is only an approximation, but if you can draw the boundary of the whole farm, it's much more accurate. Yeah, that makes sense. So you are working in Peru. You mentioned that your expertise is in rural environments. We're also talking about cocoa and coffee farmers. So usually these farmers are in rural, remote environments, obviously. And I wonder what are the challenges of these environments? Specifically? I think I will say that one of the biggest challenges is connectivity. Many of the areas where coffee and cocoa are grown are very remote and there's often no reliable Internet access. So any data collection system needs to work completely offline. Another challenge I will say is logistics. These farms are often located in forested areas that can be difficult to access. So this is one method, such as walking the perimeter of the farm or drawing its boundaries using downloaded satellite imagery become particularly important. Even if the face of the challenges, these approaches allow us to capture the full extent of the plot and ensure that geospatial data collected remains as accurate and reliable as possible. And maybe just to help our listeners actually understand what we Are talking about this IDE when an enumerator reaches a farm and it needs to start creating the boundary. I think usually what you would do with Internet connection is that you would probably use a software that would open a map. And just to simplify, maybe we can think about how we open Google Maps as we are walking in a city. But usually these maps, which contain a lot of reference points and are constantly loading, they are based on Internet connection. So it becomes essential, because you don't have this Internet connection, that your enumerators have access to a map that works well even without the Internet connection, so that they can walk around, make sure that they are in the right part of the map, and collecting multiple geopoints that together will form a shape. It will form a polygon. So, and how you were mentioning downloaded satellite images. So is this how you are collecting the data without Internet connection? Yeah, basically we prepare the maps in advance and store them directly on the tablets used by enumerators. Before fieldwork begins, these maps are downloaded and packaged so that device can access them locally without needing an Internet connection. And yeah, like you said, this is really important in remote rural areas where connectivity is often unreliable. By having these maps already on the device, enumerators can visualize the landscape, including fields, rivers, roads and nearby landmarks, which help them orient themselves, identify the correct location of each farm. We use NB tiles for these maps. I think the easiest way to think about an MB tile is just that it's an offline map, just like that. Instead of relying on Google Maps or other online map services, this map that is an MBT is already saved directly on the device. It contains the satellite imagery and map tiles needed to visualize the area. So everything works locally without having to load data from the Internet. Great. Can you Very, very briefly, because I know how technical this can get and how deep this can get. Can you briefly describe how you can use MBTiles and survey CT in the field? First, we use gistools to generate these envy tiles and this can be uploaded directly to the tablets used by enumerators in the field. Once the maps are stored in these devices, enumerators can open them within service while conducting the survey. This allows them to see satellite imagery of the landscape without Internet and do all the drawing and all the walking. In places like San Martin, where farms are often surrounded by forests and access can be challenging, having this visual reference is very helpful. It was very helpful in our project. It helps innovators all in themselves. That's very helpful, I think and for those that are wondering more about mbtiles, maybe I can just have a side note mentioning that we can link any supporting articles that we have in surveycl around mbtals in the episode notes. I think that's going to be helpful too. It is a great feature for those that want to collect these polygons without Internet connection for sure. And I think one of the advantages of mbtalys is like the integration with SAP cto. It's really practical and it's really easy to do it. So like you said, there was a webinar that I participated that we integrate these two features and I think it can escalate to another context. So it's pretty useful. And so thinking about your projects, you have collected all the data, you have been in the field collecting the data. What happens next? How does the Rainforest alliance and the farmers themselves, I don't know, use this data? Yeah, once the data collect is collected, it can be analyzed and combined with other sources of information, like satellite imagery from governments or other organizations in general. First, what I actually do is process the data using Python and goodies, which is extremely useful because it allows you to organize and visualize all the farm layers in a very practical way. You can have all the layers in one giant file. And I think this can allow organizations like the Rainforest alliance to monitor land use and access compliance with environmental regulations annually. So this can be crossed with other sources of information, maybe administrative data or satellite imagery from other sources like governments. So Besides meeting these EU Dr. The EU Deforestation Regulation, the land mapping itself seems to benefit farmers not just to access the EU market, but maybe for other policies. I don't know. What do you think? Yeah, I think one important benefit for farmers is maintaining access to international markets. And many global markets are inducing stronger sustainability requirements. And having reliable farm data help farmers demonstrate that their production practices meet the standards, particularly in Latin America where there's not much traceability. So in that sense, I will say that good data collection helps connect farmers to markets. Do you have any specific stories or examples that you would like to share about the farmers you have been working with? Yeah, I have one. I think, for example, when we were on the field, when in emers started tracing the boundaries of the farms on the tablet, farmers are often surprised to see their land represented it on a satellite image. And it's really curious because it's like they get surprised and like, whoa, yeah, that's my farm. Yeah, that's the one. And they feel confident about that. People are seeing that they are not Meeting the standards, they are compliant and they are producing their coffee and cocoa in the most correct way. I will say that makes sense. Do you feel like from your experience and you have been mostly working with the Rainforest alliance for the eudr, but do you feel like this data collection process could be applied for any other compliance or any other certification programs? Yes, definitely. I think the combination of digital service, geospatial mapping and offline tools can be applied in many different contexts. Sustainability standards and traceability requirements continue to grow. Because they will definitely continue. I think we'll see this type of approach used more widely, maybe across different organizations, governments. I will say maybe not only the European Union, maybe United States or Asia countries. It can be used more in the future. And as the demand for this high quality geospatial data grows, are you seeing any trends that are exciting to you? Anything particularly innovating? That is exciting? One trend that is really exciting is that I will say that it's integration between field data and satellite data. When you combine accurate ground level data with remote sensing, it becomes possible to monitor environmental change at much larger scales. That has huge potential for sustainability monitoring. Also something that has caught my attention lately is that there are some advances in satellite mapping. Using machine learning models developed by Google to analyze large volumes of earth observation data, this tool can automatically detect field boundaries, vegetation patterns and land use changes from satellite imagery, helping generate more accurate and detailed maps of agricultural landscapes. So yeah, I will say those two. Great choice. So looking ahead and not just thinking about geospatial data, thinking about technology in general, I feel like we see technology being implemented more and more in a smallholder agriculture. I mean for data collection like geospatial data, but also in other things like for example, farmers might be using more technology for their practices to monitor their land. How do you see the role of technology in this context? I think technology will play an increasingly important role in making smallholder supply chains more transparent, efficient and maybe resilient. Tools like digital data collection, satellite imagery and geospatial mapping are already helping organizations better understand where farmers are located, how land is being used, and how production systems are evolving over time. At the same time, these technologies are becoming more accessible. What used to require specialized equipment or large technical teams can now often be done with relatively simple tools like mobile data tablets. Collection platforms and also technology nowadays is not enough. The real value comes from building systems that allow organizations to collect reliable data and actually use it to inform decisions that include embedding in the training of the field teams, designing clear data Workflows and making sure that information collected in the field can be integrated with other sources of data. So I would say that looking ahead, I think the key step for organization is to start building those capabilities now. Developing strong data systems, investing in geospatial literacy and creating process that ensure data collected in the field translate into insights that can support farmers. Great. And I think in a way that might relate to my next question because I was going to ask you if there is an organization that is just starting to explore geospatial data collection, if you had a piece of advice to them and I think that piece of advice is already helpful. I don't know if you have any more specific to geospatial data collection. Yeah, definitely. I think I would like to highlight three things. First, it's really important to invest in training field teams. Collecting geospatial data requires more than just knowing how to use the software farm. It also involves understanding how to capture accurate boundaries, how to interpret satellite imagery and handle challenging field conditions. When enumerators are well trained and understand why the data matters, the quality of waffle collect improves significantly. Second, design workflows that prioritize data quality from the start. This means building the right checks into the data collection process such as GPS accuracy thresholds, clear protocols for mapping plots and validation rules in the survey instrument. For example, for this project we found a building function inside services deal that allow you to calculate the area of the polygon that you just draw. So that's super useful and yeah, and third, I will say that make sure the data being collected is clearly linked to how it actually be used. Geospatial data collection can be demanding in the field. So it's important that every variable and measurement serves as a has a clear purpose. When the team understands how the data will support analysis, traceability or decision making, it becomes easier to design a process that focuses on collecting information that is both relevant and actionable. So I will say those three. Thank you for those. And as the last question, do you have any specific platforms or tools or resources that you would like to share and recommend to our listeners? First, obviously SAR cto I think like it's the tool for salaries. You can do many things to SAR CTO and one of them is to geospatial data collection. It allows you to work online but also in offline settings like in rural context. Also my favorite one is Codys. I think it's a great open source tool. It allows you to work with satellite imagery, map and plots. I combine different type of layers it's pretty useful. The last one maybe like also I'm trying to familiarize more with these tools but I just give you this one. It's like Google Earth engine and data sets such as Sentinel are also pretty helpful. So yeah, we'll see rctl, qgis, Google Earth engine and many Python Python. It's also really important to know. I think it's pretty much language easy to learn, it's accessible for everyone. So I will say those well thank you so much Nicole for being here. I think it was a wonderful conversation about geospatial data collection and I hope it was very useful to our listeners too. Thank you so much Mata for having me here. That wraps up our conversation with Nicole from Letwright. Today we discussed navigating EUDR compliance and how accurate land mapping keeps coffee and cocoa farmers connected to global markets mastering offline data collection using MPTILES and Survey CTO to capture precise GPS polygons in remote disconnected areas and finally, the future of traceability by integrating ground level field data with satellite imagery and machine learning. You can check the episode notes for links to Nicole's work and resources on MBTiles. And don't forget to subscribe. Thanks for listening to Survey and Beyond the Data Collection podcast by Survey cto. If you want to learn more about how Survey CTO helps organizations collect reliable, secure and scalable data anywhere in the world, visit www.surveycto.com and if you are a fan of Survey and beyond, consider leaving us a rating or review on your favorite podcast app. Your feedback helps more listeners discover these conversations and stay connected to the latest thinking in data collection. Don't forget to follow us on Apple Podcast, Spotify or wherever you get your podcasts so you never miss an episode. On behalf of the entire Survey CTO team, thanks again for joining us and we will see you next time.