When Drones Become the New Farm Tools: Who Defines ‘Scientific Farming’?

Crashing is common, not just for agricultural drones, but for other commercial ones as well. Once a crash occurs, the drone in flight is no longer under the operator’s control; it either plummets heavily to the ground or gets caught on an obstacle.

This particular crash, which I happened to witness, left several surrounding villages without power for an entire day. As the accident was caused by a drone snapping a power line, the locals were understanding. A day’s blackout didn’t have much of an impact on rural life; the only person hit hard was the operator himself.

The operator, Mr Li, is in his forties and had been in the trade for over two years. The accident cost him dearly. With the propellers smashed, he rushed the drone to the dealer. The replacement parts were out of stock in the county town, so he had to wait several days for them to be shipped from the provincial capital. The cost of the new parts and shipping totalled less than 1,000 yuan, a large portion of which was covered by insurance.

But the problem is that the farming season waits for no one. The owner of the land found a new operator that very day to finish the remaining spraying; Mr Li had to hand over the other jobs he had booked for the following few days to other operators. Rough calculations suggest he could have earned over 1,000 yuan from this job and the subsequent ones; now, he won’t earn a single penny.

He also had to bear other losses resulting from the accident. The snapped wire was a main line supplying power to several nearby villages. Upon receiving the call, the power company immediately dispatched a team for emergency repairs. Excluding the compensation for lost work time caused by the blackout, Mr Li ended up paying over 2,000 yuan for other repair and personnel costs.

This crash happened because Mr Li failed to manually mark the obstacles. He said, “The radar on the drone can detect most obstacles, but these hanging wires aren’t as easy to pick up. Before I started, I asked Liu Er (the landowner) if there were any power lines in the field. He said there weren’t, so I didn’t walk the field to mark them on the remote. Who would have thought I’d be so unlucky!”

Generally, before operating a drone, a pilot will first plan and survey the plot. Surveying involves using high-precision sensors on the drone to take aerial photos of the farmland, efficiently gathering spatial data of the surface to set operation boundaries, mark obstacles, and plan flight paths. The intelligence of drone surveying is improving every year, and most common obstacles can now be identified and marked.

However, regardless of how much drone technology advances, there are always special circumstances that require cautious handling by the pilot; surveying cannot be left entirely to the machine. The user manual mentions that for the DJI T100 released in 2024, obstacle detection may fail for small obstacles, slanted power lines, or objects level with the aircraft’s landing gear. In such cases, the aircraft must be controlled manually to prevent accidents. Radar performance is also limited when encountering vertical poles or electricity pylons, requiring users to fly with caution. Otherwise, one slip-up—just like Mr Li and that black wire—and the drone ends up snagged on a line.

There are many causes for drone crashes, and hitting power lines is one of them. Precautions for working near high-voltage lines are covered during pilot training, and DJI Agriculture’s official website also provides solutions for such issues. Long-time operators have also summarised their own “homegrown rules”. Brother Yang, who has been a pilot for nearly ten years, has “Three No-Fly Rules”: no flying at noon (due to thermal currents), no flying over graves (due to magnetic anomalies), and no flying near high-voltage lines (due to signal interference).

But beyond operator error, technical mistakes, and environmental factors, the “black-boxing” of technology is also a significant cause of crashes. “Crashing” is a highly professional and common problem; engineers in the industry have the technical resources to study and address it, but for pilots who acquire knowledge through experience, the causes of such accidents remain a “black box”. They don’t know exactly where the machine failed and can only guess the cause based on their own experiences and stories they’ve heard.

For pilots and farmers, a more immediate problem is that only the manufacturer possesses the repair technology. The time it takes to send a unit back to the factory and have it returned is long. Since the farming season waits for no one, farmers often incur higher costs to make up for the time lost due to a crash, and the pilot whose drone crashed sometimes has to share these costs.

Yet, becoming increasingly reliant on drones, they have no choice but to passively place their trust in brands with better product guarantees.

XAG’s flagship P80 agricultural drone suffered frequent accidents in the summer of 2021. Searching for “XAG P80” on short-video platforms like Jinri Toutiao, Douyin, and Kuaishou still reveals videos of farmers from across the country showing this drone crashing.

In the lands of the Northeast, drones have long since ‘taken root’. Li Changjiang, a dealer, says that while the number of drones in the county has nearly reached saturation over the last two years, people still buy them every year. Previously, it was mainly professional pilots who bought them, but now it is primarily second-generation farmers with large landholdings. These buyers have either contracted or leased vast areas of land; they are typically men aged between 30 and 45 who are keen to try new things.

Sun Bin was among the first in the village to buy a drone, fitting the profile of the typical buyer. His family has farmed for three generations and is well-known in the village as a major grain grower. Every year, his yields are used as a benchmark; everyone says Sun Bin’s land can ‘produce several more bags of grain per mu’.

To uncover the secret behind Sun Bin’s high yields, I visited his home for a field study. That morning, Sun Bin was cleaning his drone. He is in the habit of washing the machine after every few tasks, using a high-pressure nozzle to carefully rinse the fuselage and propellers, maintaining the equipment with great care.

He told me that crops generally require fertilising four to five times during the growth period. In the past, the tools available were limited, but now, with a drone, he can fertilise according to a standard schedule. ‘It’s my own machine, anyway. When the growth stage hits and it’s time to fertilise, I just do it to promote growth! It works; it definitely works.’

However, Sun Bin couldn’t explain exactly why it worked, attributing his high yields entirely to experience. ‘I’ve spent my whole life farming; it’s all based on experience. There’s no real secret—I just farm the way others do, except I’m willing to put in the hard graft (a Northeast expression meaning to endure hardship and work tirelessly).’

Many people in the village are like Sun Bin, unable to articulate the link between drones and increased yields. To increase production and income, their original method was to rely on more chemical fertilisers—the more fertiliser, the higher the yield.

The government also issues subsidies and incentives for policies such as ‘One Spray, Multiple Stimulations‘, which have ingrained a belief among farmers that more fertiliser leads to higher yields. Consequently, they are willing to buy more fertiliser and use drones to apply it more frequently. However, fertilisers cannot increase yields indefinitely, and the soil degradation caused by over-reliance on them can actually limit production.

Nowadays, farmers are more inclined to say they rely on the philosophy of ‘scientific farming’. But what exactly does it mean for farming to be ‘scientific’?

III. Who defines ‘Science’

In interviews with farmers, I discovered that in actual agricultural production, farmers’ awareness of ‘scientific farming’ comes more from agricultural input dealers and manufacturers than from the promotional efforts of government agricultural extension departments. It is fair to say that many people derive their scientific farming techniques from these sources.

Nowadays, the agricultural input industry is gradually shifting from a sellers’ market to a buyers’ market. Previously, competition existed only between different brands of pesticides and fertilisers and their dealers; now, various new technologies, such as drones, have joined the fray. To drive sales, agricultural input dealers have become adept at ‘fancy tactics’. One evening, while strolling through the village, I heard people talking about the ‘schemes’ used by the county’s agricultural shops to treat people to dinner.

‘They (the shops) call in advance and set up tables at a restaurant to invite everyone for a free meal. After dinner, an expert gives a lecture on fertilisation, pesticide application, and field management methods. If you buy the chemicals on the spot, they give you a significant discount,’ a villager told me.

In the past, agricultural shops sold locally and relied on old customers to maintain stable sales. Now, however, many dealers are adopting multiple strategies to acquire customers: some have transitioned from selling a single category to a multi-category approach including fertilisers, pesticides, and seeds; others have expanded from a single sales channel—waiting for farmers to come to them—to multiple channels, such as supplying products and services directly to major grain growers, participating in government bidding projects, or engaging in aerial protection; some have strengthened their technical marketing capabilities, providing customised fertilisation and spraying plans when farmers come to them with crop problems, ensuring the ‘disease is cured as soon as the medicine arrives’ to win the farmers’ trust.

With the rise of drone technology, agricultural input dealers have begun partnering with drone dealers for bundled sales. They repeatedly promote the benefits of drone fertilisation to farmers. For instance, Li Changjiang, a drone dealer, frequently promotes the advantages of using drones for foliar fertiliser. He says: ‘For example, when the maize is this high, applying foliar fertiliser increases yield; it’s the same for soybeans, as it can increase protein. Once the maize rows have closed and other foliar fertiliser machines can’t get in, can you afford not to spray? Look at the farmers who had high yields last year—they applied fertiliser five or six times a year. That’s why you need to use a drone to apply foliar fertiliser more often to increase yield and income.’

Farmers worry about yield losses, or rather, they pursue high crop yields because it concerns the livelihood of the entire family. The rhetoric of ‘scientific farming’ naturally strikes a chord with them. Consequently, farmers choose to buy more chemical fertilisers and use drones to apply them to the fields multiple times a year. However, whether this actually leads to increased yields does not seem to have been scientifically verified.

Whether more frequent applications of foliar fertiliser via drones necessarily lead to higher crop yields is viewed in academia as a correlation rather than a causal relationship; yet, in the mouths of agricultural input dealers, it is presented as a certainty. They use PPTs to explain packaged scientific knowledge, such as ‘nano-permeation principles of foliar fertilisers’, to prove their point, and farmers have no way of discerning the truth from the falsehood.

IV. Old Farmers and Information Cocoons

There are also many sceptical views in the village. After actual fertilisation operations, many people are left wondering, ‘I can’t see any obvious effect. Only growth retardants produce immediate results after a while.’ Some farmers believe that applying foliar fertiliser with drones is less effective than using spray irrigation vehicles; for the same dose of concentrate, the amount of dilution water used by irrigation vehicles is about ten times that of drones. The greater the volume of water, the better the plant absorbs the nutrients; the concentration of drone fertilisation is too high for the plants to absorb.

Second Uncle has also farmed his whole life and is a staunch drone opponent, believing they are useless for increasing yields.

Beyond his 40 years of farming experience, the 60-something Second Uncle’s ‘fertilisation is useless’ theory stems from economic calculations and internet ‘experts’.

With the rise of e-commerce, buying pesticides and fertilisers via live-streaming platforms has become a new channel, greatly disrupting traditional sales models. Every day, tens of thousands of people stream on short-video platforms like Douyin and Kuaishou; among these agricultural streamers are farmers, agricultural input dealers, and some self-proclaimed agricultural ‘experts’.

Some strongly advocate for drone fertilisation to boost production and yields, while others firmly oppose it, claiming it is useless. Both sides present data, examples, and experience to support their views, making it difficult for farmers to distinguish truth from fiction amidst the chaotic information. Furthermore, the recommendation algorithms of short-video platforms continuously push accounts and viewpoints that interest the user while filtering out opposing or disliked voices. Consequently, within the farming community, ‘how the land should be farmed’ has effectively formed one information cocoon after another.

He also has an economic calculation. Previously, using spray irrigation vehicles for fertilisation meant fertilising twice a year; once the maize or soybean plants grew too tall and the vehicle could no longer enter, the process stopped. The cost of fertiliser was calculated at approximately 200 yuan per *shang* (around 15 mu). Now, using drones for fertilisation is not restricted by plant height, so fertilisation can occur throughout the growth cycle. According to the dealers, fertilising around five times a year is most effective, and the cost of fertiliser rises to over 500 yuan per *shang*.

‘Drones are said to increase yields by 15%, but with grain prices so low, can the extra three *dou* of grain even cover the cost of the pesticides and fertilisers?’ Second Uncle says.

Around the National Day holiday in 2024 was the period for large-scale maize and soybean harvesting in the Northeast. The farmers I surveyed in the summer all said this year’s harvest was very good and yields were high. Second Uncle only fertilised twice this year, and his yield showed no significant decrease compared to villagers who fertilised five or six times.

Both the staunch supporters and the opponents of drones were joyful over the harvest. However, grain prices in 2024 were far lower than in previous years. Starting from that autumn, the price of maize in the Northeast fell continuously, dropping from the usual 1.2-1.4 yuan/jin to 1.1 yuan/jin this year. This means the sales price for one *shang* of maize was at least 2,000 yuan less than in previous years. Because farmers using drones had higher costs—spending at least 500 yuan more per *shang* on fertilisers and pesticides, plus 200-300 yuan per *shang* to hire a drone pilot—the final net profit per *shang* was negligible for both the supporters and the opponents.

V. Increased Production $neq$ Increased Income: The Overlooked Link

“Increased production does not equal increased income” is a classic paradox in agricultural economics; though seemingly contradictory, it profoundly reflects the complexities of the agricultural production chain. This logical conflict is particularly evident within China’s smallholder economy.

The continuous decline in maize prices is partly due to increased market supply, which includes this year’s increased yields as well as old stockpiles from previous harvests that remained unsold. This oversupply has led to a situation where “plummeting prices penalise the farmer”. Furthermore, the rising volume of imported maize over the last two years has further saturated the supply.

Yet, as long as final yields improve, farmers are willing to invest more in costs and increase fertiliser application, driven by a mindset that is as much about confidence as it is about gambling.

It should be noted that farmers always retain the right to decide whether to use more fertiliser; they do not blindly follow the word of agricultural input dealers, but instead rely on a decision-making system rooted in experience. The most critical factor in this system is yield. If the autumn harvest shows an increase, then increasing fertiliser is deemed effective, leading to higher investment the following year. This is what Second Uncle meant by: “the better the land’s year, the more you throw these things into it.”

Conversely, a drop in yield undermines the farmer’s confidence, making investments in the following year more conservative and cautious. This reflects the inherent nature of agriculture as an industry dealing with biological entities—as the Northeast saying goes, “crops don’t yield every year” (meaning that even if this year’s harvest is poor, one must persevere and plant again next year).

Yet, it is more a matter of gambling—investing high costs to maintain stable high yields in the hope of hitting a high purchase price. After all, for the farmer, this is the only lever they have.

Meanwhile, farmers remain sceptical of new technologies: do drones truly increase grain yields? And even if yields do rise, why aren’t their pockets getting any heavier?

Faced with these questions, the fleets of drones offer no answers; on the earth, there remain only patches of neatly partitioned land—priced, yet silent.

Foodthink Author

Yvonne

A person dedicated to understanding and explaining an “Anthropocene” that has exhausted the external. A PhD candidate in Sociology, focusing on agriculture, environment, and climate change

 

 

 

About the Lianhe Creative Project

This article was supported by the Lianhe Creative Project. The author returned to their hometown in 2024 to conduct research on the use of drones by farmers in the Northeast.

Regarding Foodthink’s “Lianhe Creative Project”: In order to understand the current state of food and agriculture, and to support more people in exploring the complexities behind food and farming issues, Foodthink, together with several non-profit and media partners, launched the 2024 Lianhe Creative Project. The project supports media creators and researchers in conducting field research within the food and agriculture sector and provides funding for them to complete content creation for the public.

After multiple interviews by six judges, 18 projects were ultimately selected for support by the Foodthink Lianhe Creative Project, of which 11 have been published:

“Amei the Cleaner Wants a Proper Meal | The Worker’s Table”

“In Malaysia, Chinese Traders Only Want Grade A Durians”

“‘Fake Meat’ Ousts Real Meat: Herders, Dining Tables, and the Amazon”

“Ensuring Sweet Watermelons, Enduring Bitter Farming”

“From the Guashan Yao to ‘Chosen Mushroom Hunters’: The Termite Mushroom Craze”

“Malan in Shenzhen Has No Dining Companion”

“Why Has the Sweetness of Childhood Vanished?”

“Technology, Pesticides, and Pilots: The Other Side of the ‘Technical Revolution'”

“Who Forced the Wet Markets Out?”

“Will Fresh Produce E-commerce Make Wet Markets Disappear?”

🔽

Beyond high-cost, high-tech paths, for small-to-medium farmers with limited means, mastering ecological agricultural techniques that are locally adapted, low-cost, and easy to operate is a more realistic and viable option. From 23 to 25 December 2025, Foodthink and its partners will host the second “Ecological Agriculture Local System Co-construction and Learning Camp” in the suburbs of Guangzhou. We invite “local experts” and mentor farmers working on the front lines to provide hands-on teaching in practical techniques such as soil improvement, composting, and liquid fertiliser preparation, while exploring market opportunities and community co-construction possibilities for ecological agriculture. Registration is open from now until 14 December; half or full scholarships are available to apply for until 10 December! We welcome those who are transitioning or wish to commit to ecological agriculture to join us in finding your own path to sustainable farming.

All interviewees in this article are pseudonyms

Editors: Xiao Dan, Yu Yang