A report from the field:
by, William Giordano
In the first weekend of March, undergraduate and graduate students from approximately 30 universities participated in a “Hackathon” Hosted by the University of Maine and their America East Academic Consortium that focused on building tech tools for small farmers. Hackathon’s are becoming popular events in the tech world. They are competitive hardware and software prototyping events where teams tackle a problem or idea and develop hardware/software solutions while learning from the different approaches of other teams and professionals.
This event focused on developing IoT (short for the Internet of Things) devices that serve the needs of small and family farmers. Hardwired systems like this have been in use for years – think about a temperature probe in a greenhouse that is connected to the ventilation system, or thermostat for your boiler. However the new tech combined with new ways to cost effectively build – or hack – one’s own devices falls is line with Maine farmers DIY spirit. After building and using some of these systems for agriculture myself of the last few years, I’ve seen that when they are designed well to serve a specific purpose, they can have significant positive impact on a farm’s bottom line. The promise of improvements in quality of life through increased interconnectivity may stay in the eye of the beholder, however farmers are always on the edge of both technology and stewardship within the natural world. What come from the combination of the Internet of Things and the small farmer?
What was most notable about this event – and quite refreshing – was the cross disciplinary collaborations that ensued. Computer Information Science students and faculty were working alongside Sustainable Agriculture students. Electrical engineers alongside graphic designers, alongside faculty mentors from the Horticulture Department, Agriculture, New Media, as well as a few of Maine’s small farm innovators. These are the kinds of collaborations show promise for the kind of rural economic development that helps keeps Maine farm communities sustainable. Remarks from the University of Maine’s Provost Jeff Hecker, and the Dean of Student Robert Dana underscored notion that Maine’s Land Grant institution are strongly encouraging the cross disciplinary engagement for the benefit of Maine farmers. A welcomed notion indeed!
Consider that the rapid changes in mobile devices of the last ten years, and then consider that while in the 90’s and early 2000’s a young student who was uninterested in farm life setting their sights on a job in California’s tech boom, may within the near future be able to instead pursue an entrepreneurial rural tech career that can directly benefit the specific needs of a farm and positively impact the bottom line.
Data collection, and specifically integrating environmental data from sensors, can help back up a farmers hands on experience with data. The cost of building and developing these systems has dropped by orders of magnitude in the last decade making it feasible for a highly responsive mapping of environmental conditions, or behaviors of animals and plants that were previously only comprehensible from hands on experience in the field, or centrally located, wired sensors. With the new potential of low cost wireless systems, we are bound to see major updates to our standard methods for record keeping, decision making, and also for customized automation and control from a mobile device.
For a greenhouse producer this could include customized irrigation, heating, soil health practices e.g. an app that controls rollup sides for your season extension veggies or an instant read of available nitrogen in a soil root zone. For animal husbandry it means further insight into herd behavior and the behavior of individual animals that allows the farmer to address weakness or optimization before an expensive problem arises. Even basic soil testing systems are headed toward cheaper, more specific, real time readouts as opposed to field averages that are the industry standard. These systems are currently accessible to large scale productions with large scale budgets, but have been relatively elusive for smaller family farms.
The net effect of highly specialized and highly functional tech development for small farms is a promising improvement for our environmental and production monitoring systems that maximize every dollar invested in production and infrastructure. Combine that with the anticipation of young career seeking individuals not leaving for Silicon Valley after college but rather, for instance, returning to the family farm in Presque Isle to pursue their tech careers.
Consider that a family farm, or a cooperative of farms engaged in similar types of production – e.g. veggie farms in the 1- 10 acre type could keep on staff a tech developer who can custom engineer highly specified systems. Consider the value of tech developers who know the inner working of farm life, perhaps splitting their time with other old fashioned farm tasks. To be clear, I’m not talking of replacing farmers for AI (artificial intelligence) but rather engineering hardware and software systems that are customized to a farm operations. At the very least the goal here is relatively inexpensive, easy to build systems that help save costs on utilities, and stay ahead of environmental changes.
An added result of these data driven decision systems could help increase our metrics for measuring sustainability. As a society we all know how to look at our financial statements to measure financial success or hardship, however meaningful and accurate real time metrics for environmental health are harder to come by. As noted above, what if we understood how to notice exactly when nitrogen in a plants root zone became available to the plant. What if we noticed fertilizer leaching as it was leaving being washed from the soil root zone during heavy irrigation instead of when an environmental scientist reported it from a culmulative biological effect or sensor reading in a local pond, or estuary. How could we increase the efficiency of winter utility costs? How can improvements in metrics in these areas effect purchasing of yearly inputs? How would it effect day-to-day planning? How could they improve the value of every dollar we invest in our farm. Can technology help us understand life from the perspective of a plant or animal?
Students came to the Hackathon from across the Northeast, and it was notable that some of the Maine students came up with some of the most promising prototypes. Students from UMaine Greens – a division of UMaine’s Sustainable Agriculture Program who are growing year round, soil-based salads for UMaine Dining services, developed a cheap IoT device that connected to a servo motor which could be set to roll up greenhouse sides or unfurl a shade cloth.
Another UMaine team developed a data visualization application that would integrate many sensor data into a heat visual mapping tool for a computer or mobile device. The tool was a supplement to a IoT device that could would send data from many sensors (theoretically built into greenhouse structure) and presented a far more detailed analysis of temperature fluctuation than a single centralized thermostat could provide. Their system could also be used for soil conditions, light optimization etc..
A team from the University of Maryland Baltimore County developed a soil probe prototype equipped with an array of sensors that would clip on a field workers belt and at the push of a button send soil condition data to a farms central computer where it was graphed for real time and historical analysis.
Another team developed a software only database tool that “scrubbed” the websites of seed companies and organized seed varieties from the most popular seed companies into a centralized database that would allow buyers to compare details such as cost, germination rates, and traditional seed packet details on a single google-like web search with direct link to purchasing from suppliers.
The entire event underscored some of the work that has been underdevelopment through the Sustainable Year Round Agriculture project. As we work with farmers and researchers around the state to understand the ways renewable energy, heating, ventilation and lighting systems could improve farms bottom lines, we have been able to integrate these ideas into sensor based data mapping tools in ways that would have been cost prohibitive a decade ago. That One promising out growth of the SYRA project has been a longer term mobile app research and development group called the Maine Agriculture App (MAgApp) project which, like the hackthon, has been a cross disciplinary project that brought together student and faculty researchers from Agriculture, Engineering, Life Sciences and New Media across UMaine campuses (UMaine Orono and Umaine Presque Isle). To the MAgApp team the hackathon was an ideal test bed for some of the key concepts under development.
Where as a hackathon is a rapid 24 hour event, Maine Agriculture Apps (MAgApp) research and development project is building mobile apps to aggregate sensor network data for Maine farmers operating food and horticulture operations in season extension and year round greenhouses currently in use on Maine farms and at UMaine system research stations. The pilot app provides real time and historical data visualization for sensors and IoT devices developed through SYRA.
Primary goals of the pilot tool are:
- improving the metrics (visualization and accessibility) for improved efficiency in production and utility costs for farmers
- improve data driven decisions based on highly detailed environmental analysis
- create cost effective tools designed for small diversified family farmers that run intensive operations with modest budgets
Additionally when connected to IoT devices the mobile app can be utilized to make physical changes to a production systems based on data. (e.g. changing a irrigation schedule based on customized soil moisture sensor network, or greenhouse heating and ventilation based on temperature probe network data). Special care is being taken not to replace farming tasks with AI or AR systems but rather combine useful environmental analysis with farmers traditional perennial wisdom of observation to create biological feedback tools that improve a small farms work flow and the value of labor, while also providing earlier detection of potential issue that would otherwise damage crops and reduce yields.
As we all grapple with the way we can address sustainability in a changing climate, and we understand that the improving the way we work with efficiency of every task without is a key step in a transition away from fossil fuels, such tools offer an analysis that can supplement the physical presence and traditional record keeping of the farmer. But then we have to ask how to design our quest for efficiency without sacrificing quality and traditional knowledge.
This balance of needs shows promise for Maine leadership but also underscores the ways that the traditional work of Maine’s family farmers can improve a conversation and research about technology and highlight and address the past failures of rural economic development. After all similar tech systems have been applied to National Security, and NASA for decades, and billions of dollars have gone into their development in these sectors. Without a debate about what may have happened if Maine agriculture had been prioritized in lock step with these sectors all along, what now do those more concerned with national security via the effectiveness of their ploughshare and its effect on the community and local economies add to the tech sector? If I could highlight one key takeaway from this work over the years and especially this past weekend it’s that both Maine’s young people, and our experienced farmers have something equally critical to add to the conversation.
William Giordano serves as the Education Director for the Sustainable Year Round Agriculture project for the Maine Sustainble Agriculture Society.