GIS in agriculture has been boosted by the general advancement of technology in the past few decades.
The use of GIS in agriculture is all about analyzing the land, visualizing field data on a map, and putting those data to work. Powered by GIS, precision farming enables informed decisions and actions through which farmers get the most out of each acre without damaging the environment.
Speaking of tools, geospatial technology in agriculture relies on satellites, aircraft, drones, and sensors. These tools are used to make images and connect them with maps and non-visualized data. As a result, you get a map featuring crop position and health status, topography, soil type, fertilization, and similar information.
There are several applications of geoinformatics in agriculture. Let’s have a look at some of them.
From this article, you’ll learn about the following applications and use of GIS in agriculture:
- Crop yield prediction
- Crop health monitoring
- Livestock monitoring
- Insect and pest control
- Irrigation control
- Flooding, erosion, and drought control
- Farming automation
Crop yield prediction
Accurate yield prediction can help governments ensure food security and businesses forecast profits and plan budgets. The recent development of technology connecting satellites, sensing, big data, and AI can enable those predictions.
One of the most profound techniques in this field is Convolutional Neural Networks (ConvNets or CNNs). A ConvNet is a deep learning algorithm that is taught to identify the productivity of a crop. Developers train this algorithm by feeding it images of crops whose yield is already known to find productivity patterns. CNN has an accuracy of about 82%.
A crop-prediction technique workflow
Source: Sustainability and Artificial Intelligence Lab, Stanford University
Crop health monitoring
Checking crop health across multiple acres manually is the least efficient option. This is where remote sensing combined with GIS in farming comes to the rescue.
Satellite images and input information can be paired to assess environmental conditions across the field, such as humidity, air temperature, surface conditions, and others. Based on GIS, precision farming can upgrade such an assessmen and help you decide which crops require more attention.
A more sophisticated approach uses imaging sensors on satellites and air vehicles to check the temperature of crops. When the temperature is above normal, this might indicate a disease, infestation or insufficient irrigation.
Neural networks like CNN, Radial Basis Function Network (RBFN), Perceptron, and others can be helpful in assessing crop health too. The algorithms can analyze images for unhealthy patterns.
Livestock monitoring
The simplest application of farm GIS software in animal husbandry is the tracking of movement of specific animals. This helps farmers find them on a farm and monitor their health, fertility, and nutrition. GIS services that allow you to do that comprise trackers installed on animals and a mobile device that receives and visualizes information from those trackers.
Here’s one example. You want to monitor the weight of your beef cattle. Each animal has a tracker on its ear or neck. Every time it steps on the digital scales, the scales read the ID of that animal and assign a new value to that ID in the system.
You don’t need to manually enter that data. Meanwhile, if there’s an alarming change in the animal’s weight, you can quickly find that animal and check its health.
There are also more interesting use cases of farm GIS software, such as preventing wolf-cattle encounters. There are ambiguous spatial specifics that affect the distribution of wildlife in an area, including wolves. We could reduce undesirable encounters by understanding those subtle specifics, which could be done by the combined use of AI and GIS in agriculture.
Insect and pest control
The invasion of harmful insects and pests, or infestation, does heavy damage to agriculture. A look from above can enable accurate, timely alarms to prevent that.
Yet even high-resolution images might not provide visible early signs of infestation.
The alternative would be using AI. You develop a neural network and train it using deep learning algorithms. Through this training, you feed the neural network images of infested land, and the network learns to find samples that indicate infestation. After that, you feed it satellite images of the land you want analyzed.
As mentioned above, you can also use remote sensing along with geospatial technology in agriculture to check the temperature of the crops. Plants respond to infestation by heating up as they stop getting enough water or nutrition.
Irrigation control
Keeping an eye on vast fields to make sure that each crop gets enough water is a challenging task, but one easily tackled by geoinformatics in agriculture.
Aircraft and satellites equipped with high-resolution cameras take images that allow AI algorithms to calculate the water stress in each crop and spot visual patterns behind water shortages.
Pair those images with water delivery system maps, and you will find out how well your current irrigation scheme is performing.
Flooding, erosion, and drought control
Marrying GIS and agriculture can help prevent, assess, and mitigate the negative impact of destructive natural phenomena.
To identify flood-susceptible areas, you can use flood inventory mapping techniques. You need to collect data such as past floods, field surveys, and satellite images. Use those data to create a dataset to train a neural network to spot and map flood risks, and you will create an ultimate disaster management tool.
If you need to check land for susceptibility for soil erosion, you could pair Universal Soil Loss Equation (USLE) with GIS and remote sensing. Run satellite images through spectral analysis to check USLE factors and verify those images with field observations. As a result, you can create a map featuring the level of deterioration of the soil across the field.
Similar GIS solutions for agriculture can be used to control drought.
Farming automation
Seeding machines, intelligent irrigation systems, driverless harvesters, and weed remover robots are the inevitable future. You could equip each of your machines with sophisticated sensors, but why do that if you can connect them to an integral GIS system?
(That is not to say that automated vehicles don’t need sensors — they do.)
GIS in farming can provide precise maps, including all necessary information about the crops in the field. Maps like those are called task maps or application maps. Smart machines use them to tend to the field.
Here’s an example of how GIS solutions for agriculture might work. Once a GIS system has detected weed infestation, it assigns a “Weeding needed” label to that area. The weed remover robot reads the label and places this area on its list of tasks.
Apart from providing signals for machines, task maps can help unskilled workers do their job more efficiently.
Conclusion
If you browse the Internet for use cases of GIS in agriculture, you will find articles and studies dating back to the early nineties. The objectives of agriculture haven’t changed much since then — nor have the problems that GIS is expected to solve.
However, technology has made a huge leap in the past three decades, and now is the best time to look for new solutions to old problems. Agriculture software development will help the industry feed the world, cut food prices, and save the planet.
