Drones as the New Farm Tools: Who Defines ‘Scientific Farming’?
I. Crash Behind the Black Box
Crashes are commonplace, affecting not just agricultural drones but other commercial models as well. The moment a crash occurs, the drone is instantly beyond the pilot’s control, either plummeting into the ground or snagging on an obstacle.
The drone crash I chanced upon left several neighbouring villages without power for a full day. As the accident was caused by a drone severing a power line, the locals expressed understanding. A day without electricity hardly disrupts rural life. The only person to suffer severe consequences was the pilot at fault.
Mr Li, who is in his forties and has been operating drones for just over two years, suffered heavy losses from this incident. With the propellers damaged, he immediately took the drone to a dealer. The replacement parts were unavailable in the county seat, so he had to wait several days for them to be shipped from the provincial capital. The cost of the new parts, including shipping, totalled just under 1,000 yuan, with insurance covering the majority.
But the real issue was that agricultural timing is unforgiving. The landowner hired another pilot that very day to complete the remaining spraying, and Mr Li was forced to reassign his other booked jobs to colleagues. A job that should have netted him roughly 1,000 yuan, when combined with the following days’ work, now yielded nothing at all.
He also bore the other costs stemming from the crash. The severed cable was a main line powering several nearby villages, and the local power utility dispatched an emergency repair crew as soon as they were notified. Factoring in repair and labour costs, but excluding compensation for lost income caused by the blackout, Mr Li ultimately paid out over 2,000 yuan in compensation.
The crash happened because Mr Li neglected to manually mark the obstacles. He explained, “The drone’s radar can detect most obstacles, but suspended cables like this are notoriously difficult to register. Before starting, I asked Liu Er (the landowner) whether there were any power lines in the area. He said there weren’t, so I skipped walking the field myself to map out the operational zone on the controller. Who knew I’d be so unlucky as to hit one!”
As a standard procedure, drone operators map and survey the target plot before beginning work. This surveying process utilises high-precision sensors mounted on the drone to capture aerial imagery of the farmland, efficiently collecting spatial data of the ground surface. Operators then use this data to define operational boundaries, mark obstacles, and chart flight paths. The intelligence behind drone mapping has been advancing steadily, enabling the system to identify and tag most conventional obstacles.
Yet, regardless of how far drone technology advances, there will always be edge cases that demand a pilot’s caution; mapping cannot be left entirely to the machine. The user manual for the DJI T100, launched in 2024, warns that obstacle detection may fail when dealing with thin objects, diagonal cables, or items aligned with the aircraft’s landing gear. Pilots must be prepared to take manual control to avoid accidents. Radar performance is also limited when facing vertical poles or transmission towers, requiring extra vigilance. A single lapse in concentration can result in a scenario just like Mr Li’s: the drone snagging onto a black cable and being left hanging.
There are many reasons drones crash, and striking a wire is just one of them. Precautions for operating near high-voltage cables are covered in pilot training, and DJI Agriculture’s official website also provides guidance on how to handle these hazards. Pilots who have been operating for years have also distilled their own practical field rules. Mr Yang, who has been flying for nearly a decade, adheres to a ‘three no-flys’ principle: no flying at midday (thermal updrafts), no flying over graveyards (magnetic anomalies), and no flying near high-voltage lines (signal interference).
Beyond operator error, technical mishaps, and environmental factors, the opaque, black-box nature of the technology remains a significant technical cause of drone crashes. Drone crashes represent a highly specialised yet commonplace issue. While industry engineers possess the technical resources to diagnose and mitigate them, pilots who rely on hard-earned experience find the underlying causes just as opaque as a “black box”. Lacking visibility into exactly what has failed within the machine, they are left to speculate on the cause, drawing solely from personal encounters and hearsay.
For pilots and farmers, the more pressing issue is that repair expertise resides solely with the manufacturers. The turnaround time from sending a unit back to the factory to receiving it fully serviced can stretch on for weeks. The farming calendar does not pause. Farmers frequently incur additional costs to make up for lost time, and the pilot responsible for the crash is sometimes asked to shoulder a portion of this burden.
Yet, as their reliance on drones grows, they have little choice but to place their trust in brands that offer stronger product warranties and support.
XAG’s flagship P80 agricultural drone was plagued by accidents throughout the summer of 2021. A search for “极飞P80” on short-video platforms such as Toutiao, Douyin, and Kuaishou still yields footage of farmers from across the country documenting crashes of this model.
XAG’s after-sales team ultimately proposed a compromise: a paid buyback scheme for the mapping system; an additional ¥20,000 in insurance coverage for users in Heilongjiang; and a commitment to free repairs for the spreading system within a stipulated timeframe that year.
Three years on, the shadow cast by this incident still lingers over the farming community. Among the dozen pilots and farmers I interviewed, every single one exclusively uses DJI equipment. Zhao Mingliang, a prominent grain producer in the village, purchased a new drone earlier this year. When I asked why he hadn’t opted for XAG’s latest P150, given that both models are priced similarly and offer comparable capabilities, he replied, “For one, XAG suffered far too many crashes in the preceding two years, and the fallout is simply too costly to bear. For another, their local after-sales support is sparse, making repairs a nightmare when things go wrong.”
This new breed of agricultural machinery, heralded as a flagship of smart farming, evokes a complex love-hate relationship. When it comes up in conversation, farmers often vent their frustration with a muttered curse: “Bloody thing, you can never work out what’s wrong with it!”
Yet, they remain utterly dependent on them. Deploying a drone holds out the promise of higher yields, though, if you stop to think about it, the logic behind it can still feel a bit like a mystery.
II. The High-Yield Anxiety in the Fields
Sun Bin was among the earliest in his village to purchase a drone, fitting neatly into the emerging buyer profile. His family has farmed for three generations, making him a renowned large-scale grain producer locally. His annual yields are constantly held up for comparison, with neighbours saying his fields yield “several extra bags of grain per mu”.
Hoping to uncover the secret behind his high yields, I made a trip to his farmstead. That morning, Sun Bin was cleaning his drone. It is his habit to wash it after a few flights, carefully rinsing the fuselage and rotors with a high-pressure nozzle and attending to the machine with meticulous care.
He explained that crops typically require fertilising four or five times during the growing season. Where traditional tools once imposed limits, the drone now allows him to apply fertiliser according to standard agricultural schedules. “It’s my own machine, after all. Whenever the crop hits a key growth stage, I head out and apply the fertiliser to boost development. It’s effective, absolutely effective.”
Yet, when pressed for specifics, Sun Bin struggled to explain the mechanics behind it, attributing his high yields entirely to experience. “I’ve been farming all my life. It’s all down to experience. What secrets could there be? I farm just like everyone else. It simply comes down to being willing to put in the hard graft (a regional turn of phrase meaning ready to endure hardship and work tirelessly).”
Like Sun Bin, many villagers struggle to articulate the precise link between drone use and increased yields. In their quest for higher yields and income, they previously relied on a simple formula: apply more chemical fertiliser, reap a greater harvest.
The government has also rolled out policy subsidies and incentives, including programmes like “one spray, multiple benefits”. These initiatives have reinforced a prevailing mindset among farmers: more fertiliser equals higher yields. As a result, they are keen to purchase larger quantities of chemical fertiliser and deploy their drones for more frequent applications. Yet, chemical fertiliser cannot drive yields upwards indefinitely; moreover, the soil degradation wrought by such reliance ultimately acts as a brake on productivity.
Today, farmers are far more inclined to claim that their methods are grounded in scientific farming. But exactly what does it mean to farm “scientifically”?
III. Who is defining “scientific”?
Today, the agricultural input sector has gradually shifted from a seller’s market to a buyer’s market. Competition was once confined to rival pesticide and fertiliser brands and their retailers; now, new technologies such as drones have joined the fray. To move stock, input dealers have become adept at marketing ploys. One evening, while strolling through the village, I overheard locals discussing the standard playbook county-level agricultural supply shops use when treating customers to meals.
“They’ll call ahead, book a room at a restaurant, and host a free meal for everyone. Afterward, an expert delivers a talk on fertiliser application, pesticide spraying, and field management. If you place an order on the spot, you’ll get a substantial discount,” a villager present told me.
Previously, agricultural supply shops relied on local proximity and loyal repeat customers to maintain steady sales. Now, however, dealers across many regions are employing a variety of customer acquisition strategies. Some have shifted from selling a single product category to offering a broad range of fertilisers, pesticides, and seeds. Others have expanded beyond passive, in-store sales, branching out into direct supplies for large-scale grain growers, bidding for government tenders, and offering drone crop protection services. Still others are bolstering their technical marketing skills: when farmers bring them crop problems, they design customised fertiliser and spray programmes to ensure the issue is swiftly resolved, thereby winning their trust.
Following the rise of drone technology, agricultural input retailers have started collaborating with drone dealers to offer bundled packages. They consistently promote the advantages of drone-applied fertilisation to farmers. For example, drone dealer Li Changjiang frequently advocates for using drones to spray foliar fertiliser. He explains: “Take corn at this height, for instance. Applying foliar fertiliser boosts yield, and it’s the same for soybeans, where it raises protein levels. Once the corn canopy closes over the rows, conventional sprayers can’t navigate the field—so aren’t you going to apply it? Look at the farmers who achieved high yields last year: they applied foliar fertiliser five or six times a season. That’s why you need drones to make multiple foliar applications, increasing both yield and income.”
Academic research suggests a correlation, not causation, between increased drone spraying of foliar fertilisers and boosted crop output. Yet agricultural supply dealers treat it as a guarantee. They pitch their claims using PowerPoint presentations on packaged-up scientific concepts like the ‘nanometre penetration principle of foliar fertilisers’, leaving farmers unable to judge the truth for themselves.
IV. Veteran Farmers Caught in an Information Cocoon
Uncle Er, who has spent his entire life farming, is a staunch opponent of drones. He maintains they do absolutely nothing to boost yields.
Alongside four decades of farming experience, Uncle Er’s sixty-something years have shaped his “fertiliser-is-pointless” stance, which he backs with a ledger of costs and the words of internet “experts”.
The rise of e-commerce has made live-streaming platforms a new channel for purchasing agrochemicals, severely disrupting traditional supply chains. Now, tens of thousands of people broadcast daily on short-video apps such as Douyin and Kuaishou. Among agricultural streamers, some are actual farmers, others are agrochemical dealers, and a few simply label themselves agricultural “experts”.
Some loudly champion drone fertilisation as a driver of productivity and higher yields, while others staunchly oppose it as useless. Both camps marshal data, anecdotes, and experience to back their claims, leaving farmers struggling to separate fact from fiction amid the noise. Meanwhile, short-video algorithms continuously feed users the accounts and opinions they prefer, while filtering out dissent. As a result, when it comes to the question of “how should land actually be farmed?”, farmers find themselves trapped in ever-tightening information cocoons.
His scepticism is also grounded in hard economics. Traditionally, with sprayer trucks, farmers would fertilise twice a year. Once the maize or soybean plants grew too tall for the truck to navigate through, applications would stop. Calculated this way, fertiliser costs ran to around 200 yuan per *shang* (15 *mu*). With drones, height restrictions vanish, allowing applications throughout the entire growth cycle. Dealers recommend roughly five applications a year for optimal results, pushing the fertiliser bill to over 500 yuan per *shang*.
“They claim drones boost yields by 15 per cent, but with grain prices this low, will the extra harvest even cover the cost of agrochemicals?” Uncle Er asks.
Around the National Day holiday in 2024, Northeast China enters its peak harvest window for maize and soybeans. Farmers I spoke to this summer all report an excellent season with high yields. Uncle Er applied fertiliser just twice this year, yet his harvest shows no meaningful drop compared to neighbours who applied it five or six times.
Both the staunch drone proponents and the sceptics alike celebrated the bumper harvest. Yet 2024 grain prices languished far below previous years. From that autumn onwards, maize prices in the Northeast slid continuously, dropping from the usual 1.2–1.4 yuan per *jin* to just 1.1 yuan this year. Consequently, the selling price for a *shang* of maize fell by at least 2,000 yuan compared to past seasons. On top of that, drone adopters faced higher overheads, spending an extra 500 yuan per *shang* on fertilisers and pesticides, plus 200–300 yuan to hire the drones themselves. When all is said and done, a careful reckoning reveals that whether you champion or shun drones, the net profit from farming a single *shang* has all but evaporated.
V. Increased Yields ≠ Increased Income: The Overlooked Chain
The sustained decline in maize prices stems partly from an oversupply in the market. This surplus comprises both this year’s expanded harvest and unsold stock from last season, creating a classic case of “cheap grain hurting farmers.” Compounding this, the steady rise in maize imports over the past two years has further flooded the market.
Yet, so long as the harvest ultimately improves, farmers are often willing to stretch their budgets and apply heavier doses of fertiliser, driven by a mindset that blends confidence with calculated risk.
It is important to note that farmers retain full autonomy over whether to apply additional fertiliser. They do not blindly follow the recommendations of agrochemical suppliers; rather, they rely on an empirical decision-making framework rooted in experience. Yield remains the decisive factor. If a bumper autumn harvest justifies the extra input, they will scale up their investment the following year. As Second Uncle put it: “The better the land performs one year, the more inclined you are to pump these inputs into it.”
Conversely, a drop in harvest dampens farmers’ confidence, prompting a more cautious and restrained approach to spending in the following season. This underscores a fundamental trait of agriculture as a biological enterprise: the cycle cannot be paused. As the saying goes in the Northeast, “You must keep planting crops year after year” (a dialectal phrase meaning that even after a poor harvest, the fields must be planted again next season).
At its core, however, this reflects a gambler’s mindset: committing to high input costs to sustain stable, bumper harvests in the hope of striking lucky with favourable grain procurement prices. After all, for the average farmer, this is often the only lever they have left to pull.
Meanwhile, scepticism towards new technology persists among farming communities: Do drones truly boost grain yields? And even if they do, why do farmers’ pockets remain empty?
The drone swarms offer no answers. All that remains across the landscape are neatly demarcated plots—priced, yet utterly silent.
About the Lianhe Creation Project
About Foodthink’s “Lianhe Creation Project”: Designed to shed light on the present state of food and agriculture, and to empower more voices to explore the complexities underpinning these sectors. Foodthink, alongside several charitable and media partners, launched the 2024 Lianhe Creation Project to back media creators and researchers conducting fieldwork in food and farming, providing grants to produce public-facing content.
Following multiple rounds of interviews by a panel of six judges, 18 projects were ultimately selected for support through the Foodthink Lianhe Creation Project. Eleven have been published to date:
“A-Mei, the Cleaner, Just Wants a Decent Meal | The Worker’s Table”
“In Malaysia, Chinese Buyers Only Want Grade A Durian”
“‘Fake Meat’ Ousts the Real Thing: Herders, Dining Tables, and Amazon”
“Sweet Watermelons, Bitter Farming”
“From Mountain-Dwelling Yao to ‘Chosen Foragers’: The Termite Mushroom Craze”
“Ma Lan in Shenzhen: No Dining Companion in Sight”
“Why Have Childhood Sweets Vanished?”
“Tech, Pesticides, and Drone Pilots: The Flip Side of the ‘Technological Revolution'”
Rather than pursuing capital-intensive, high-tech solutions, small and medium-sized farms with limited financial reserves would find greater practical value in mastering ecological farming techniques that are tailored to local conditions, low-cost, and straightforward to implement. From 23 to 25 December 2025, Foodthink, alongside our partners, will host the second ‘Local Ecological Farming Co-Creation & Learning Camp’ on the outskirts of Guangzhou. We will invite seasoned grassroots specialists and experienced farmer mentors to provide hands-on training in practical skills such as soil conditioning, composting, and liquid fertiliser preparation. The programme will also explore market prospects for ecological produce and opportunities for community-led agricultural initiatives. Registration is open now until 14 December, with applications for partial or full scholarships accepted until 10 December. Whether you are currently transitioning to sustainable practices or looking to embark on ecological farming, we invite you to join us in charting a viable, sustainable path for your own land.
All interviewees referred to in this article are identified by pseudonyms.
Edited by: Xiaodan, Yuyang