Agriculture is one of the world’s oldest industries, with evidence of farming dating back as far as 21,000 BCE. However, population growth and environmental disruption caused by climate change require agriculture to find new solutions to improve food security.
One way to produce more food is to use data analytics to improve agricultural efficiency. By analyzing data on weather patterns, soil conditions, pest infestations, sunlight levels, market forces, and other factors, farmers can determine the best times to sow, process, and harvest crops, as well as which crops are best to grow in specific situations to maximize profits.
Making decisions based on data is nothing new to agriculture. Since 1792, the Old Farmer’s Almanac has provided American farmers with information on weather forecasts, crop charts, and astronomical data, while in the UK, Farmer’s Weekly provides regular sources of information and updates to assist farmers. These publications and similar have provided regularly updated information to enable sound decision making.
Giving farmers more data and advanced tools to analyze it allows them to make more informed decisions about how to manage their crops. There is already a huge amount of historical weather data that can be used to predict weather patterns. Climate change is likely to increase the likelihood of extreme weather, but such phenomena are built into existing climate models.
There are many sources of data relevant to agriculture: stock market data can be collected to determine which crops are doing better than others, sensors can be used to collect data on soil quality and moisture, drones can be deployed to monitor for pests and weeds, and satellites can be sent into orbit to collect data on the amount of sunlight in fields.
WWF and TechUK recently published ” Coding crops: the role of data in promoting sustainable agricultural trade and responsible supply chains “, a report that highlights the role of data and technology in promoting sustainable agricultural practices and responsible supply chains globally.
The study highlights the importance of using mobile technologies and digital platforms to monitor data at the production level to ensure supply chain visibility and sustainability. Additionally, the authors provide strategic recommendations to the UK government to scale up technological innovation.
The report concludes: “Data harmonization offers numerous opportunities, from enabling farmers to better understand the impacts of their production and the supply chains they are part of, to supporting financial institutions to make informed decisions and verifying claims of sustainable production and distribution.”
Finding labour on farms is a big problem, so farmers are open to using technology in any form. David Ross, from the Scottish Rural College
One of the main benefits of data analytics is the ability to continuously monitor incoming data over a period of time and send alerts when problems arise – for example, sensors can identify the best times to irrigate or fertilize crops.
Data analytics systems can predict likely outcomes by identifying trends within existing data. Forecasting solutions such as Monte Carlo simulation can help farmers identify market trends and growing conditions for different crops.
“We’re now collecting data not just to support decisions around growing produce, but also in the realm of forecasting,” said Mark Wolff, advisory industry consultant and chief health analytics strategist for the global Internet of Things (IoT) division at SAS.
“If we were to do this combination of these things at this time, based on certain assumptions about irrigation, what should we expect?”
The trials and tribulations of farming
Computer simulations can be used to predict what would happen if farmers grew different crops or tried new routines. Previously, farmers had to trial grow a crop for a season to get a first-hand idea of how profitable it would be, which was time-consuming and could lead to losses.
“The next level of analysis is the digital twin, or simulation of the workflow,” Wolf says. “Once we know mathematically the relationship between the inputs and outputs (the crop, the genetic makeup of that crop in a particular area, the set of chemical and biological inputs, and the outcome), we can simulate that.”
Money isn’t the only consideration in agriculture. Farmers often choose crops for reasons other than immediate financial gain. Crop rotation is a common agricultural technique in which different crops are grown in rotation to control pests and weeds and improve the soil.
“Rather than telling you what crops to grow, there is scope to identify outbreaks of disease in parts of a field, like an early warning system,” says David Ross, principal consultant for arable services at the Scottish Rural College.
In all of this, economic feasibility is key. Very few farmers can afford a comprehensive data analytics software suite. It may be a one-time payment, but the payback period will be long. Then, as the product becomes outdated or unsupported, more significant investments will be required.
For these reasons, agricultural technology companies are exploring the practicality of offering data analytics services to farmers as a subscription service, which would overcome the issue of high upfront software costs.
However, investment in a range of sensors will still be necessary.Many agricultural companies are already generating large datasets that can be used as baseline models for agricultural data analysis.
This will require a data collection network, which combines sensors to collect data into a network that transmits the information for storage and analysis. Installing such a network can involve large one-time costs, as can the process of connecting businesses to communications services.
Overcoming connection issues
Additionally, connectivity remains an issue in rural areas, especially in remote parts of the country. Some rural areas still lack mobile phone coverage. For these data analytics solutions to be effective, further investment in communications infrastructure in rural areas is needed.
One potential challenge for analytics-driven agriculture is the lack of information sharing between different providers. There is little to no interoperability between devices from different manufacturers, so farmers have lots of different data sets but no way to combine them and see how they interact with each other.
“Farms use dozens of different pieces of software to input and output data, and I call them data silos,” says Ed Harris of the Centre for Data Science Agriculture Research at Harper Adams University. “They don’t talk to each other, and they only give you information about the farm.”
Being able to combine multiple data sets provides a more comprehensive overview of a farm’s operations, rather than a series of separate data entries. However, protecting the personal and financial information of each farm whilst sharing this data is essential to remain compliant with data protection regulations such as the UK’s Data Protection Act 2018.
Being able to compare data with other farms gives farmers a better understanding of how their farm is operating. This information has little value on its own, but when combined it allows for a baseline estimate of how a typical farm is operating. This can then help farmers identify areas on their farm that need to focus their attention.
“What I don’t particularly like is that if you get those numbers from one farm, you only have one number for that farm,” Harris says. “I think what’s needed is a business case for benchmarking carbon and other farm operations so we can get a little more information about how other farms are doing.”
There is also a regional factor to consider: a data analytics system developed using farms in one part of the country will not necessarily work in other parts of the country.
Each data analytics solution needs to adjust for environmental and geographical issues to be effective. “It worked perfectly when we worked on a project in Australia, but when we brought it into Scottish conditions, the environment is different and it was a disaster,” says Ross.
In tandem with data analytics, automation is being used to address recent labor shortages: these are typically repetitive and monotonous tasks such as milking cows, allowing workers to focus their attention on more complex tasks.
“Getting labour on farms is a big issue so farmers are open to using technology in any way they can,” says Ross. “A robot that costs £30,000 a year on a dairy farm could save a farmer £85,000 a year in labour.”
Although agriculture is not considered an IT-driven industry, it is one that requires reliable and accurate data to enable informed decision-making. Data analytics can be used to identify emerging trends within datasets and alert farmers when interventions may be needed to protect crops and increase yields.
“Farmers are cynics. It has to be proven and have a return on investment,” concludes Ross. “It’s not necessarily cash, it might be time. It has to be relatively simple, robust and work on dirty farms and in dirty environments, because there’s no getting around it.”
