How AI is modernizing Chinese agriculture

Since ancient times China has been an agricultural country, with both huge demand for and large-scale production of a wide range of agricultural products. China’s per capita arable land area however is far less than the world average, and the quality of superior arable land is relatively small. According to figures from the World Bank and the National Bureau of Statistics, in 2016, China’s cultivated land area was 135 million hectares, accounting for 8% of the world’s total cultivated land; while the total population was 1.38 billion, accounting for 19% of the world population. Also, the farmland quality rate is low, which affects food production and the quality of agricultural products. Moreover, China’s agricultural sector is often seriously affected by natural disasters — causing average annual crop damage of nearly 20,000 hectares and grain production losses of 30-40 billion kilograms. This has motivated China to continuously develop new agricultural technologies.

 

In recent years frontier technologies such as AI, big data, IoT, and 3S technology (RS remote sensing technology, GIS geographic information systems, and GPS global positioning systems), have been aggressively deployed to accelerate the modernization of Chinese agriculture. These new technologies are being applied mainly in planting, animal husbandry, and agricultural services.

 

 Comparison of China and World cultivated land area per person
 

Cloud farming: high-tech assistant for precision agriculture
 

Modern precision agriculture techniques are applied to both small-scale and large-scale farming in China. The MAP (Modern Agriculture Platform) platform developed by China’s largest agriculture input enterprise Sinochem Agriculture is operational in seven provinces with varied weather and geography for the planting of the wheat, corn, and other crops. The system provides tailored seed suggestions before the planting season, and weather forecasts with a focused accuracy of one square kilometer to help with sowing, spraying, watering, and harvesting. The optimal type and amount of fertilizers and pesticides can also be calculated automatically corresponding to crop varieties and other considerations. Moreover, farmers can check real-time humidity and temperature conditions, and even receive pest and disease warnings on a smartphone via the MAP App.

 

Technical services for farmers
 

New technologies can also help by improving technical services. The Agricultural Technology Center of Kailu government in northeastern Neimenggu has developed a smartphone app for end-user technical services that provides local weather forecasts and remote observations, agriculture and animal husbandry E-education, pest and disease diagnosis, and expert video consultations. Farmers can also use the app to remotely observe the state of their crops, and for example to send photos of a discovered pest or disease to get feedback or suggestions. It’s also possible to make an appointment with an expert for a video consultation via the app, saving farmers from visiting a town or city for such services. Overall the app aims to provide farmers with convenient, efficient, and timely solutions.

 

Remote sensing and drone assisted agricultural insurance
 

AI technologies are also being applied by many Chinese insurance companies for on-site damage investigation and determination and to improve efficiency and reduce costs. In northwestern Ningxia for example, the Science and Technology Department has deployed a classification and extraction insurance software for local specialty crops Goji, melon, grape, and jujube. The remote sensing information model was developed by studying the relationship between the hyperspectral data of plants and domestic high-grade satellite images. The support vector machine based classification methods use disaster distribution surveys and other data as a basis for risk assessment and premium determination. The technique can reduce insurance investigation costs and streamline the claims process.

 

 Distribution of specialty crops in Zhongning, Ningxia. Blue: jujube; yellow: melon; red: Goji.
 

There are still a number of obstacles to the widespread deployment of AI technologies in agriculture. Although AI can bring many improvements and advantages, many farmers have taken a wait-and-see attitude, mainly due to the initial investment costs and doubts regarding the robustness of these new technologies.

 

The Chinese government has introduced measures to promote smart agriculture, starting with encouraging development of AI, 5G, and IoT to provide better technical solutions. Smart agriculture has been designated part of the national development strategy; new research institutes and databases for smart agriculture have been built; and more and more local governments are cooperating with research institutions and enterprises to provide informative lectures and technical support. The government also provides incentives such as tax breaks, loans and subsidies to encourage farmers to adopt the new agricultural technologies. China believes these policies and technologies can build a bountiful future for the nation’s farmers.