Most of the big data in agriculture started with satellite telemetry. Since satellite telemetry data in many countries are open and standardized, some newcomers in the field also start with satellite telemetry and local meteorological data. Although the foundation of big data in agriculture has been established, due to the insufficient industrialization of domestic agriculture and lack of commercial agricultural insurance in some emerging countries, the big data in agriculture based on satellite telemetry has been unable to find a scenario of application and fail to create value.
In terms of data sources, at present, most of the big data in agriculture rely on specialized and industrialized farms. The data sources are not limited to satellite telemetry, but cover more channels, including professional unmanned aerial vehicle(UAV) and small crop protection UAV. The data accuracy of satellite telemetry is about 5m; the data accuracy of professional UAV is about 1m; and the data accuracy of small crop protection UAV is 3-10cm. Data sources have grown accurate and diversified.
In terms of implementation, big data in agriculture basically involves the utilization of satellite data and meteorological data for data calculation, analysis, visual display and in-depth development on the basis of Google map or Google Earth. These data are mainly used for crop yield monitoring and prediction, planting field planning and analysis, actuarial pricing of agricultural products and so on. The big data in agriculture made by some Chinese institutions rely on planting, spot trading and other data to analyze the output and price of agricultural products and display the dynamic market.
In terms of data collection, more and more tools can be used for collecting data in agriculture. In addition to commercial satellites and geographical data collection, there are also public sector UAVs, crop protection UAVs, robots for agricultural production, new models of data monitoring and collection equipment on agricultural machinery, and various cameras with different scope of accuracy and technology in farmland, greenhouse and on agricultural machinery and UAVs, all kinds of sensors, especially in smart agriculture sector, and all kinds of drip irrigation equipment.
Big data in agriculture can be applied in many scenarios with various fields of data.
* Satellite telemetry and mapping
Satellite data and meteorological data can provide reference for crop yield prediction and pest prevention.
* Land data
In addition to the land data telemetered by satellite, the data of soil and fertilizer stations, the promotion data of formula fertilizer by soil testing and the data of land leveling and drip irrigation equipment can be collected and supplemented by UAVs and other tools in the process of production and crop protection.
* Planting data
They are collected through field management services and trusteeship services in the whole process of agricultural production, including planting varieties, production processes, fertilization, pesticide use and harvest.
* Crop protection data
They include agricultural technology, agricultural chemistry, agricultural machinery and other data in crop protection.
Through the crop monitoring data and crop protection data collected by agricultural UAVs in the process of pollination, fertilization and pesticide application, plant protection and pest detection, as well as the prediction and analysis of pests and diseases and yield prediction
* Agricultural product data
They can be collected by monitoring the growth process of agricultural products, sorting and detection in fruit size, sweetness, acidity or starch, pesticide residue, metal residue and logistics tracking of agricultural products, electronic tags of agricultural products.
* E-commerce data
Transaction and sales data of e-commerce of agricultural products.
A large number of scattered and fragmented data related to agricultural products on social media and the Internet will be collected.
From the perspective of industrialization, the core of big data in agriculture are asset and transaction data. Assets include mainly land and agricultural products while transaction mainly refers to the business flow, logistics and information flow of agricultural products transactions.
What are the application scenarios of big data in agriculture?
However, if big data is not created applicably for agriculture sector, it’s meaningless. We must collect, utilize and analyze data based on algorithms in combination with the actual application scenarios of agriculture.