Agriculture is ultimately an ancient science that’s been iterated upon for generations. With each new crop, each new family, and each new culture, we’ve discovered more about optimal farming techniques. Due to this, we generally understand agriculture from a localized perspective – we understand planting seasons, soil types, and other variables within the farming process have profound localized impacts on yield, flavor, quality, and more.
Unfortunately, however, this data is not always something that can be extrapolated to a global perspective. The global farming industry is very different from your average subsistence farm or cooperative field, and because of this, unique data is needed for farmers to bridge their understanding between global practices, lessons, knowledge, and localized efforts.
This is exactly what high-tech data APIs offer the average farmer. Using APIs in farming is a high-tech solution to a low-tech problem, and offers a great deal in solving problems of efficiency, efficacy, and long-term maintenance of the farming industry from mega-farm to local cooperative.
Types of Data Collected by Farming APIs
When we discuss data-driven APIs for farming, it helps to look at what type of data is actually being used. While APIs can deal with a wide ranging set of data, most farming APIs are concerned with Global Data, Regional Data, and Local Data.
Global Data is the broadest of these data types, and contains sales information, international shipping route guidance, spoilage rates, and more, ultimately functioning as a frame of reference for import and export. This type of data is predominantly used by APIs that concern themselves with reducing loss to spoilage, ensuring compliance with international trade regulations, and even with domestic shipping in non-contiguous landmasses (such as Alaska, where some shipping occurs through an intermediary country despite the fact that both the source and destination of crops are in fact domestic). This type of data is also key for financial and transactional APIs, as a stout auditing trail must be established in such cases.
Regional Data is still quite broad, but is less concerned with issues of international trade and more with specific regional issues. Weather tracking, collective farming data, water table information, and more are key data points that have a huge effect on crop management and collective regional farm work. These types of APIs are typically of a collective nature – i.e., farms provide localized data that is then collated across many data points to provide a collective overview of the region at large. These APIs also tend to depend heavily on municipal and regional government data points, such as drought reports, levy health, and other factors handled specifically by state and local governments.
Local Data is specific to the individual farm. CO2 statistics for the soil, crop performance indicators and prediction engines, planting and harvesting dates, and other highly localized information are typically found in this space. While Global Data and Regional Data is often a collective endeavor, localized data processing is usually offered through a combination of Internet of Things devices and some sort of System as a Service offering.
Collectively, the use of any combination of these data systems results in a farming “cell” of sorts, allowing the farm to operate as an individual, unique, and self-contained farm while utilizing data and metrics generated by the greater farming community. Because farmers can use as much or as little of these data sources as they want (or are willing and able to pay for), the approach to data in farming is extremely granular, with each specific use case demanding its own custom implementation.
The Need for Agricultural Data
We’ve identified the types of data these APIs might use, but there’s a very simple question left to be answered – what is the value of this information? Why move to use data at all, given how effective farming has been historically?
Simply put, farming is a balance between efficiency and production, with an added concern for sustainability and long-term field health. Waste is, in fact, one of the biggest loss leaders in a farm, and as such, waste must be reduced. The problem is that waste, whether it be through poor planting methods or spoilage reduction, is hard to measure, hard to control, and hard to prevent. This is exactly where data-driven agriculture comes into play.
Waste can be reduced through collective efforts, which largely rely on data generated by willing participants. This data can inform proper planting methods, effective harvest cycles, more efficient use of the water table, and more – ultimately, this data can help create a situation of lower waste and less spoilage, which ultimately results in greater amounts of product and revenue.
Ultimately, there are two ecosystems of sorts – a natural ecosystem where produce is created, and a digital ecosystem of producers. Utilizing data is key to ensuring efficiency and efficacy in both ecosystems, improving efficiency, reducing cost, lowering waste, and generally making the network of producers better at what they do.
Agricultural APIs vs Generic APIs
A point should of course be made that when we discuss farming APIs, we’re talking about a very specialized, very specific type of API. Generic APIs can be useful in some applications, but in terms of agriculture, attempting to use them for farming purposes is essentially a square peg and round hole situation. Farming data is very specific, and as such, a generic weather API or municipal data source isn’t really appropriate, as they tend to offer a single purpose, unspecialized look at the data on hand.
Agricultural APIs utilize agriculturally-specific data sets – from these sets, they add value to one another by working collectively. Weather data and their resultant models are only so useful – when presented together, however, these data sets morph from pure data into usable, actionable information.
Simply put, commercial and generic APIs are fine, but standardized, agricultural-specific APIs are going to deliver more bang for the buck for farming, maximizing success, profit, efficiency, efficacy, and reducing waste through specific, related data.
When talking about such a broad concept, it often helps to look at some specific use cases for these types of APIs. The following four case studies show exactly how this data is useful, and why it is poised to become ubiquitous in the coming years.
Case Study 1 – Fujitsu and Aeon’s Hà Nam Province Trials
Fujitsu and Aeon, both massive juggernauts in the Asian marketplace, launched a field trial using a series of interconnected devices and some agriculture-specific APIs. The runaway successes of these ICT techniques, especially in consolidating and utilizing data from sources including farm-work records, growing conditions, and cultivation environment, inspired the field trials in Vietnam to see if the same success could be replicated.
The study tied data from disparate sources, including from local smartphone operations by farmers and field workers, and aggregated the data to serve as agricultural guidance. The data specifically tested a range of farming techniques to find hidden links between field health, technique, planting schedules, and their relationship to business elements of cost, revenue, and quality.
This is a perfect example of how the joining of localized and regional data in a business-centric API can serve to leverage existing data sources to greater heights. While much of this data already existed, being able to collate those data sources and see their relationships are key to improving international trade and the business functions of producers.
Case Study 2 – Iteris ClearAg
Whereas the Aeon Vietnam field trial is more an example of a generalized benefit of APIs in this space, ClearAg is a more specific example of the value of specific agricultural microservices. Iteris ClearAg offers several unique APIs, each with their own specific feature set designed to deliver actionable data for precision farming.
The Field Weather API utilizes sensor data and in-house Iteris meteorology staff to generate data on precipitation, hail, frost, wind speed, wind direction, temperature, and more, all with the goals of pointing towards optimal planting and utilization on a field by field basis.
The Crop Health API utilizes sensors and collected data as well as models for nutrients, drydown, growth, and more to monitor and improve crop health.
These are only two of their APIs – additional offerings such as the Map Overlay, Soil Conditions, Account, and ClearAg components each offer likewise complex and diverse functions, delivering actionable data for farmers.
Case Study 3 – Agworld
Of course, not all agricultural APIs are going to be centered upon growing processes. Some APIs and solutions are going to be tangentially related to agriculture, offering tertiary functions that may not directly tie to the farming world, but are of absolute value to those within the industry. A great example of this is Agworld. Agworld is a cloud-based platform with a JSON API that delivers read-only cooperative data on farms and agriculture businesses. They then use this data to form a collective understanding amongst willing parties.
This is great for planning to-market transactions, timing collaborative grows to fulfill large orders, rotating fields and leasing space to other farmers. Even collaboratively distributing resources from a common pool could be achieved. Ultimately, Agworld’s API drives premium offerings which branch these functions into other, different directions, but the API itself is a freemium offering that promises some major functions for very low cost.
Case Study 4 – National Data and Analytics
While the previous examples we’ve discussed are quite limited in scope, a great example of a more wide-ranging example is the CropScape API. The CropScape API utilizes the National Agricultural Statistics Service from the USDA to deliver some very powerful data to farmers and agricultural workers.
The API ties into geo-referenced crop specific data, and then overlays this data over satellite imagery to give regional and national data references. This data can then be used to not only deliver more efficient processing for farmers and related workers, but also deliver commercial data to large-scale markets and distribution centers, identifying areas of potential shortages and excesses.
The API has both RESTful and SOAP variants, and delivers this data in a wide range of data formats, including text, CSV, and JSON. This data could be used to cross-reference against other APIs and data sources, showing cropland field data and creating a framework for not only crop health on a national scale, but planting patterns, behavioral data for regions of consumers, and even the effect of historic, real-time, and predicted weather patterns on the national farming industry.
Precision agriculture software is definitely a new concept, but it’s the end result of decades of development in connected, collaborative farming. As more time passes and additional methodologies are discovered to improve the technology of farming in the modern age, IoT data is only going to be that much more important and impactful.
By adopting data-driven agriculture, farms decrease waste and improve management. The simple fact is that farms and governments are already providing these data sets. Not using them is simply allowing for this data to go to waste when it could be leveraged for your own successes.
We’ve just scratched the surface of a thriving industry with exciting developments in sensors, IoT, drones, and robotics. APIs are at the heart of this transformation. Did we leave anything out? Please comment below!