 This is a story about a computational tool that benefits farmers and society. So nitrogen is an essential nutrient for crop production. And about 100 years ago, a process was invented for the manufacturing of synthetic nitrogen fertilizer, which then became widely adopted in agriculture and has facilitated big yield increases and really prevented a lot of widespread hunger around the world. But there are negative consequences to the use of nitrogen. It's very energy-intensive to produce, therefore expensive to the farmers, and also it has multiple environmental impacts. Greenhouse gas emissions through nitrous oxide and water quality concerns causing groundwater contamination and hypoxia problems in estuaries like the Baltic Sea, the Gulf of Mexico, and even the Yellow Sea near here. One of the key issues is that farmers tend to over-apply due to uncertainty about what the optimum end rate is and due to risk. They're trying to avoid risk of under-fertilizing their crops. We were trying to find a solution to the balance between the agronomic benefits and the environmental needs and designed a tool basically that would allow for precision nitrogen management recommendations. So the challenge was it was a very complex problem. Reactive nitrogen is very dynamic in the environment. It has many sources and lost pathways and is highly influenced by the production environment. Weather, soil, and management. So we envisioned a tool that was cloud-based with excellent communication capabilities including mobile devices. It would be highly computational, data-intensive, and it would be true to the scientific principles. We had some challenges ahead. First of all, we had a very unconventional idea that was very difficult to fund even within the research environment, and we needed a very diverse and multi-talented team. We rolled out ADAPTEN as we called it in 2008 but realized it needed to be commercialized, which we accomplished in 2013 through a startup company. So a quick overview of the tool. It has multiple software models and it uses some user inputs and then it dreams near real-time weather data to provide daily updates. The outputs are multiple, including recommendations on a 20 by 20 meter spatial resolution if that's desired. We learned about some key needed features. One of them is even though we have a fairly complex system, the user experience needed to be simple and straightforward. The other one is that it's important to integrate with other farm management software so that you would expand your clientele rapidly and could scale the product. We also built in a lot of transparency, so the user gets recommendations, gets answered, but he or she also gets support information, including even some irrigation recommendations that go with that. And that helps build user confidence. So let's not forget about data privacy and security. It's a topic that I've already heard a couple of times today. It's very important to U.S. farmers. They also highly value the independence of recommendations, not associated with any product sales. So the company secured it in a highly secure environment and also established a Grower Bill of Rights to explain the privacy policies. We need to prove the technology in the field. And so we did over 200 experiments on farms and were able to prove that, yes, it's a win-win proposition. You can save the farmer money and you can reduce the environmental impacts. So that's the optimum outcome. There's still some barriers. This is fairly disruptive technology to the industry that is generally fairly non-transparent and is still very much focused on moving product. Even the research community took a little while to get used to this. But I think that there are many opportunities, basically, to benefit the farmer, to benefit society, and even to benefit the fertilizer industry. We've established sustainability credentials and there's several supply chain management programs now that are interested in this tool, including Walmart's Global Sustainability Initiative. The challenge was the complexity of the problem and you needed a computational approach and a highly data-intensive approach. We needed the right people. We needed a multi-talented team and it was difficult to fund an out-of-the-box idea. But we were able to find the right commercial partners that were still independent of product sales. So some final thoughts. Computation can tackle complex sustainability problems and can be implemented in the field. Science-based approach is best, but commercialization is essential. So my question that I leave with you is are there more such win-win opportunities?