This is the second article in a series of articles aimed at understanding digital agriculture and be guided in adopting it. In the previous article, we had an Orientation to Digital Agriculture
In this article, you can find a general structure and definitions of digital agriculture understanding how words such as the Internet of Things, big data, cloud, robots,… fits in the big picture of digital agriculture.
Digital Agriculture Step 1: Listen
This is the first step in Digital Agriculture: data acquisition. There is a lot of talk about data and…for good reasons as without data (= information) how could we take any decision?
Digital Agriculture gravitates around data! The key concept here is that digitizing data we are making them available for easy processing. It’s sort of as grain and flour. Imagine flour as “digitalization” of grain. Once you have flour you can easily process it to make bread, cakes, pasta, anything. Easy to combine, store,… So an easy way to think about data is considering them as ingredients to prepare a recipe or consumable material to be used for building or fixing, using tools (tools comes next in the “processing” phase).
So… the first step of digital agriculture starts with Data Gathering, what for?
Plain English: know your present conditions (at any time)
Digital Agriculture jargon: “CONTINUOUS DATA GATHERING”
Or…in other words:
Plain English: let your “digital friend” knowing as much as possible about your agronomic situation
Digital Agriculture jargon: It’s a dynamic assessment of your crops, soil condition, weather patterns, commodity prices, available genotypes, major threats, water availability,…
This is nothing new in agronomic practices (as in any other domain), what makes the difference in digital agriculture are innovative enabling technologies that allow introducing efficiency in data gathering and especially getting data that our previous agronomic colleagues were just dreaming of.
Thinking about data as the “ingredients for food” than, these technologies are like the “producers of such ingredients”.
What are these enabling technologies? I’ll present them here, in short, some of you may be familiar with some of them, anyway in coming articles I’ll explain better the roles of each of them, while here they are just shortly introduced. What’s more important is to place these technologies into this phase of digital agriculture, as this will help you to get the big picture of Digital Agriculture:
- IoT (internet of things: sensors gathering regularly soil humidity, air temperature, air humidity, leaf wetness, solar radiation)
- Satellites (providing low-resolution images on a regular base)
- Drones (imagery at higher resolution)
- Smartphones (enabling operators to take notes and pictures)
- Precision farming machinery (tracking actions: Smart Ploughing, precision seeding, treatments,…)
- (Robots… coming soon, some are already there!)
Simple rules to follow:
Now, there are several good recommendations to follow when implementing the above enabling technologies to adopt digital agriculture. While experts can guide you in each specific domain, 2 rules should be always with you:
- Bad data → poor (or wrong) decisions (“junk in junk out”): although technology reliability is improving at tremendous speed, never, ever rely 100% on what are called “raw-data” (meaning data coming directly from a sensor, a drone, ….), but give yourself (your system) always the “benefit of the doubt”. What does it mean? Think about the flour example again. If the flour is spoiled, how would a cake you made taste? In direct terms: god decisions shouldn’t be taken from single datasets, but from pre-processed and combined set of data that statistically improve the overall assessment while “curing crazy data” (that happens, believe me).
- More data → better decisions. This is also common sense, the more ingredients you have the more choice to make the perfect recipe, right? The more information about your field, the higher the probability you would know how much & when to irrigate, or estimating the real risk of disease to come, or guess the right time for the harvest to maximize the BRIX,… but…
- More data → not necessarily = better decisions. Ok, I don’t want to confuse you, but also common sense tells you that if you intend to prepare a cake probably 1kg of pepper is useless to you. What I’m trying to say is that often there is a frenzy to gather data (especially pushed from technology vendors) that sometime won’t translate in value (= investment without ROI).
Digital Agriculture Step 2: Think
This is the second step in Digital Agriculture: data processing. Going back to the “ingredients (data) for recipes”, data processing is like “cooking”. As we don’t eat flour it needs to be processed to become something edible. Similarly, data needs to be processed to become “recommendations“, from which we can extract value taking the best possible decisions helping to reduce unnecessary time and material while reducing as well agronomic risk and optimizing harvest. These are the ultimate goal and motivations behind any adoption of digital agriculture.
What does “processing” mean?
Plain English: compare your situation with your objectives and available resources to take convenient decisions.
Digital Agriculture jargon: “data processing to transform digital data into reporting, recommendations, optimizations, and predictions“.
As each of us is able to think (processing information) why do we even need digital agriculture support in processing? Because Digital Agriculture data gathering is powerful and the data gathered are tremendously more than in the past therefore we risk ending up in a situation where…
Plain English: too many data are confusing, your “digital friend” supports you extracting recommendations to take better agronomic decisions
Digital Agriculture jargon: discovery, interpretation, and communication of meaningful patterns in data
Luckily this critical step is also supported by innovative enabling technologies that may fall under the Digital Agriculture technologies. The number and diversity of such technologies (along with the ones mentioned in the data-gathering section) confuses even most technology experts, so don’t be scared if you feel lost, for the time being, just have in your mind that the following technologies are enabling this second step in several ways.
- Cloud database (data gathered stored in a secure, accessible & safe place)
- Analytics (set of digital tools helping to “digest data” (process data) to extract information/recommendations)
- Models & Algorithms (set of rules often expressed in mathematical or informatic form; they are nothing more than a way to combine data to make assessments and take decisions, we daily use them in our lives without even noticing)
- Artificial Intelligence (higher computer autonomy in interpreting data to provide recommendations)
Rules to follow:
As for the previous step, is a good idea to focus and keep in mind a few basic rules:
- Mind the “all-in-a-box” dream solution(s): several “solution-in-a-box” for digital agriculture offer attracting solutions that seem as simple as plug&play while solving that specific problem. While often they can really help there is an underlying problem that can emerge afterword when you start piling on many of such solutions working independently from each other and quite often being NOT INTEROPERABLE. But sooner or later “all chickens come home to roost” and you will end up in a situation where your digital friends may at best not leveraging each other, at worst they can make conflicting decisions.
- Mind the “silos thinking trap”: in addition to the above, always keep in mind the “big picture”, meaning looking a problem from different angles (different data types & sources), because focusing on sub-set of data while overlooking the big picture may lead to big mistakes;
- Leverage local agronomic experience: often the solution is already available (the local agronomist already know how to improve water efficiency provided he/she has the data…), it just needs to be leveraged by digital technologies; open solutions (unfortunately still not significantly present on the market) will enable this opportunity soon;
Digital Agriculture Step 3: Act
This is the third step in Digital Agriculture: taking a decision and executing accordingly. All previous steps are just a setup to guide in taking the best possible action: when, what, how, how much. Of what? Everything: it may be seeding time, genotype selection, pruning time, irrigation pattern, harvest time, fertilizer dosage, pesticide treatments,… All the agronomic actions have a relevant impact on the farm costs and the overall harvest achieved.
Digital Agriculture is a combination of technologies and processes to enable the best timely decision to optimize all inputs (including work).
Therefore after having gathered data, combined them, processed them, it’s time to:
- make the agronomic decisions based on a mix of experience & precision agriculture recommendations
- act accordingly
- track actions starting the cycle again with data gathering
I put “track actions” in bold as this latest step is overlooked by most Digital Agriculture solutions and even theory, while it’s crucial to enable a virtuous cycle (*).
Going back to the “ingredients and cooking metaphor”: acting is “cooking” while “tracking actions” is writing exactly what has been done during cooking in order to know if something has to be changed to make the best recipe or just to repeat it if it was good. Well… that’s more or less, the metaphor stretches a little the reality, but it helps to understand.
More advanced readers would prefer the following: (*) in other words “tracking actions” is critical to implement the Adaptive Management Cycle in Agriculture. The latter is a quite established approach, Digital Agriculture will provide its maximum benefits only when fitting into the Adaptive Management Cycle.
- costs optimisation (time&material)
- higher profits
- reduced environmental impact
- risk mitigation
Some technologies for this step are already on the market, while many more are coming shortly. I’m convinced the sector will change dramatically with a level of automation that is currently beyond the imagination of the majority of farmers. But if you want to have an idea about the possibilities just visit most of the industrial factories where robots are managing most of the production cycle. It will happen soon in agriculture as well, starting from the large extensive farms and intensive production in greenhouses:
- irrigation adjusted to the real-time situation
- next generation tractors
- APPs: task assignment (what to do), training (how to do), execution tracking (what has been done)
- drones targeting treatments to specific plants/area
- ROBOTS herbicides targeted spraying, automatic harvest,…
1. Technologies are just tools, don’t get too excited, don’t get scared: many people are scared by robots or overexcited while they are just tools and we are used to them without even considering it; an example? We use washing machines at home for decades; while recently our homes (and cities) are becoming more and more filled up with automation, the same will happen in farms, very shortly. Don’t get scared and don’t get overexcited, always…
2. Adopt technologies according to their proven ROI adjusting to your specific farm
How Does it Work? (basic readers can skip this section)
Digital Agriculture is a process that should follow the Adaptive Management Cycle while making the best use of the most appropriate digital technology innovations to fit the specificities of the local farm. (see the coming article for an implementation plan)
Without further text, you can focus on the following image to grasp what Digital Agriculture can do for you. A careful observer can focus on this image as it tells it all.
Marco is a Digital Agriculture international expert with 20Y+ experience in leveraging digital innovations to the benefit of the market. He owns a Math degree & MBA, and before focusing on digital agriculture, he successfully worked with his teams to develop and bring to market several new technologies and products in the fields of environmental monitoring (low-cost air quality city monitoring), risk mitigation (Unesco Petra site) focusing on Internet of Things (IoT), advanced sensors, Big Data, predictive analytics, Artificial Intelligence.
He co-founded 3 companies receiving international recognition by the European Enterprise Network (innovation success in 2007), was mentioned in Forbes in 2008, was awarded the Stanford University “Best Startup Award” at the Italian Innovation Day in 2011, and was IBM Smarter Planet finalist and Global Entrepreneur.
Recent tangible achievements: saving of water up to 50% with increased production of 250% in semi-arid climate; 30% average pesticide reduction in orchards along with agronomic risk reduction.
Main customers/partners supported in Digital Agriculture: Nestlé, Syngenta, Netafim, Omya, Agroscope, Purdue University.