Viridis — Reversing Climate Change Through Technologically-Powered Farms
Our world is nearing its end.
Global warming is taking its toll on civilization. Temperatures are increasing and sea levels are rising. Entire cities are at risk of becoming submerged.
Yet, industries continue to recklessly spew greenhouse gases into our atmosphere. Billions of tons of gases like carbon are released into the air, bringing our world even closer to impending doom.
To prevent the world from a global warming-induced collapse, immediate action is necessary. The presence of greenhouse gases in our atmosphere must be curtailed as soon as possible.
Cleansing our atmosphere of huge amounts of unwanted greenhouse gases sounds like a Herculean task. At first, our team at Viridis thought that such a large-scale task would be impossible as well.
But we didn’t want to sit around and do nothing. We figured there had to be some way to prevent the world’s end.
Through pursuing the goal of eradicating climate change, we at Viridis were able to arrive at a moonshot solution that leverages Internet of Things (IoT) technology, artificial intelligence (AI), and regenerative agriculture to halt global warming for good while helping farmers make data-driven decisions.
Regenerative Agriculture — The Key to Reversing Climate Change
Carbon is a huge part of our overall greenhouse emissions, accounting for 76% of human greenhouse gas emissions. Reducing the amount of carbon in our atmosphere is a huge step in reversing climate change.
Luckily, regenerative agriculture is really good at reducing atmospheric carbon!
Simply put, regenerative agriculture is a type of farming in which plants absorb large amounts of carbon from air and store it in soil. The carbon-enriched soil becomes extremely fertile and allows for the growing of high-quality crops.
This process of absorbing carbon from the air and storing it in the soil is called carbon sequestration. To effectively sequester carbon with regenerative agriculture, some specialized farming techniques are needed.
An example is crop rotation, where, in each growing season, farmers rotate crops between fields. Throughout a growing season, crops will completely exhaust the nutrients in the soil. Planting a new crop in that field can re-enrich the soil with nutrients, allowing plants to store even more carbon in that field.
Another technique is the application of compost to crop fields, which also replenishes soil and allows crops to absorb and store carbon with greater efficiency.
These farming techniques are effective independently, but when they come together, plants become carbon-absorbing machines. The fusion of these farming tricks to accelerate carbon sequestration is what makes regenerative agriculture so powerful.
Now imagine millions of acres of these carbon-absorbing plants working together. If enough farms adapt a regenerative agriculture system, we can sequester 100% of the carbon that humans emit.
There’s just one problem, though. While these farming techniques are quite easy to adapt, farmers are still reluctant to convert their farms to a regenerative agriculture system due to a steep learning curve.
Regenerative Agriculture + IoT + AI = Superpowered Farms!
Farmers are extremely used to conventional farming methods. They’ve worked well for decades and thus they’re reluctant to convert to an unfamiliar system.
But we think that this knowledge gap can be closed with the power of AI.
We at Viridis propose the use of an AI farming assistant that can help farmers follow a regenerative agriculture system. With climate data and advanced soil sensors, an AI can be trained to tell farmers exactly when to rotate crops and apply compost to fields, making regenerative agriculture much easier to use.
Our advanced soil sensor can provide our AI with critical information on the nutrients of a farmland’s soil. By providing real-time information of soil composition, the AI can make informed recommendations on how to implement regenerative agriculture’s farming techniques.
A Full-Stack Soil Sensor
Knowing the nutrients in soil is extremely important for farms to function well. However, current methods of determining the nutrient concentrations of soil are inefficient and expensive.
Let’s say we wanted to determine the amount of carbon in soil. Figuring this out would require the use of a process called dry combustion, where a soil sample would be burned and analyzed using complex chemistry equipment.
Many farms aren’t equipped to perform such complex analysis. They have to ship their soil samples to chemistry labs and wait for long periods of time before getting results.
With our smart sensor, though, the amount of carbon in soil can be determined quickly and inexpensively.
Our sensor would be stuck into farmland soil. A process called soil visible near-infrared spectroscopy (NIR for short) is then utilized to determine the carbon content of the soil.
In soil visible NIR, infrared rays are shot at a soil sample. A spectroscope would measure how much of these infrared waves were absorbed by the soil, creating a soil absorption spectrum. AI algorithms have been developed that can then estimate the soil’s carbon contents based on its absorption spectrum.
This process doesn’t only have to be used for carbon. The same framework can be applied to get accurate readings of nutrients like nitrogen as well.
To ensure that our soil sensor is not overwhelmingly large, the spectroscopes and infrared lights would all be integrated into our sensor’s body as nanosensors. We aim to have nanosensors that read five key elements: carbon, nitrogen, phosphorus, sulfur, arsenic, and water. This gives our AI all the information it needs to provide educated regenerative agriculture recommendations.
Communicating these soil nutrient readings to our AI requires our sensor to have internet connectivity. Microcontrollers with Wi-Fi chips are integrated into each sensor, allowing the sensor to publish all of its readings to a database. The microcontroller would be powered by a small solar panel on the top of the sensor.
All of these components come together to make our super soil sensor. With the data that the sensor gathers, our AI can become the perfect regenerative agriculture assistant!
Creating Our AI Regenerative Farming Assistant
Humans are really bad at accurately predicting information.
This goes for farmers as well. It’s often difficult for a farmer to determine exactly when to rotate their crops or apply compost to soil. That’s why we at Viridis propose that we create an AI that can do this for us!
Our AI will be able to output specific dates for farmers to rotate crops and add compost. The AI can even be extended to help farmers decide when to harvest their crops.
To train an AI that can accomplish such a feat, we need an AI algorithm that’s really good at learning information. Thus, we decided to train a neural network to output accurate, informed recommendations for farmers.
But, uhh… what is a Neural Network?
A neural network is a type of AI algorithm that’s really good at prediction and classification. Often, they’re visualized through a graphic of interconnected circles called nodes.
The nodes on the left-hand side are collectively called the input layer. Here, information is inputted into the neural network. In our problem, this information would be quantities like soil nutrient data from our sensor, temperature, and other farming-related data.
The middle section, or the hidden layer, allows a neural network to develop a more in-depth understanding of the information. The larger a hidden layer (or the more nodes it has), the more complex information it can learn.
Finally, the output layer on the rightmost side returns the (you guessed it) outputs. Our problem involves outputting specific dates on which crops should be rotated, cover crops should be planted, or fields should be covered with compost.
Each of these nodes are stored in a computer as numbers called parameters. The end goal of the neural network is to learn parameters that are really good at predicting when farmers should either add compost, rotate crops, or plant cover crops. It does this by scanning large amounts of data.
Data goes through the input, hidden, and output layers (in that order) to create outputs. The model then looks at its outputs and sees how incorrect its predictions were, changing its parameters to generate more accurate predictions.
Now, how does this help regenerative agriculture?
With a neural network that can predict when farmers should take certain actions, farmers can make much more educated, data-driven decisions that sequester huge amounts of carbon. Our neural network can scan a dataset of climate and soil data to get really good at creating these decisions.
Our team created a sample dataset that our neural network would process. The dataset would contain information about the soil nutrient concentrations, which would come from our smart sensor. It would also include local climate data, like temperature and rainfall.
A quick note — these values are completely random. This is merely a sample dataset that simulates what a real dataset would look like. However, with our smart sensor and real climate data, our AI can undoubtedly be trained to make regenerative agriculture more accessible and effective!
Our Current Technological Hurdles
While we at Viridis believe that our sensor and AI combo can be extremely effective, we still have some technical barriers that need to be surmounted before the solution can become a reality.
Mainly, we lack the technology needed to build miniaturized spectroscopes.
Current spectroscopes are big and bulky. They aren’t nearly small enough to condense into a nanosensor.
We intend to bring these spectroscopes to the nanoscale, making them billions of times smaller than the current standard spectroscope.
This is an extremely ambitious goal, but we believe that it is entirely possible. In fact, institutions like MIT have already developed nanosensors for detecting elements like arsenic. And similar carbon, nitrogen, and phosphorus-detecting nanosensors are only a few years off.
At Viridis, we are concentrating all of our efforts towards creating miniaturized spectroscopes that return accurate readings of soil nutrient concentrations. Once we achieve this goal, we are confident that we can spearhead an agricultural revolution, converting farms to regenerative agriculture and putting a complete stop to global warming.
Viridis is a moonshot company created by me (Ayan), Chendur, Maharshi, and Matt with the goal of employing regenerative agriculture to simultaneously fight climate change and improve farming.
Thank you for reading to the end! Don’t forget to leave a 👏 as well! Connect with me on LinkedIn if you want to talk about AI or any emerging technologies.
