Cargill is deep in exploration mode for alternative sources of fish feed, a sprawling quest with projects as varied as creating a new protein product and growing omega-3-enriched canola on the Montana prairie.

Long a leader in land ­farming, Cargill Inc. is now growing its influence in underwater farming or, as the industry calls it, aquaculture.

The global appetite for fish is growing faster than the human population. And while fish farming has helped facilitate this rise in demand by making more available to consumers, the industry is facing several biological challenges that could limit future growth. Cargill, one of the world’s largest industrial companies, has made several recent investments aimed at solving some of the barriers.

People are eating more fish than ever before as ­science repeatedly shows such health benefits of consuming certain fish, as child development, brain functioning and immune systems. Cargill has been in the fish feed business for a while, producing feed for tilapia, catfish and other species. In fall 2015, Cargill made its biggest bet on the future of aquaculture when it bought Norwegian salmon fish feed maker EWOS for $1.5 billion, the second-largest acquisition in its 150-year history.

Wild-caught fish production has plateaued in recent years at around 90 million tons a year. Meanwhile, about 74 million tons of farmed fish is consumed annually. It amounted to around 44 percent of fish consumption in 2014, up from 26 percent in 1994 and 7 percent in 1974, the U.N. Food and Agriculture Organization reported.

EWOS, based in Bergen, Norway, controls about one-third of the world’s salmon feed supply. Consumers continue to flock to the pink fish, considered a powerhouse of health. But salmon require a very specific diet of protein and fat in order to grow and exhibit healthy characteristics that make it a desirable human food.

Traditionally, a healthy salmon diet has come from eating other smaller fish, such as anchovies or sardines, which are chock-full of omega-3 fatty acids — essential polyunsaturated fats that the human body doesn’t produce on its own. These ­little fish are what deliver the omega-3s from algae to the salmon, which then ferry the nutrients to humans.

But fish meal and fish oil, which is produced through the harvesting of these smaller fish, is putting a strain on oceans and marine life. Industry demand for fish oil will bypass the world’s fish oil supply in 2019, according to projections by Cargill’s EWOS. This is why the company is working on alternative sources of the protein and the omega-3 fatty acids, particularly EPA and DHA, found primarily in ocean life.

With links to a reduction in heart disease and other conditions, the dietary demand for these two types of fatty acids is expected to grow.

“Cargill stood back and said what do we need to do to create some sustainable platforms for this growing demand for fish. Can we create it as a sustainable supply,” said Lorin DeBonte, assistant vice president of R&D for Cargill’s ­specialty edible oils.

In November, the company announced it had developed a new type of canola seed that contains these omega-3 fatty acids and is piloting a project in Montana. Cargill spliced the EPA and DHA-producing gene from algae into the canola seed.

“EPA and DHA are not found in land plants and that’s what it makes this really unique,” DeBonte said.

Unlike livestock feed, the formulation for fish feed is highly specific and has fewer ingredients that can be substituted for the same outcome, said Einar Wathne, the former chief executive of EWOS who is now president of Cargill’s aqua nutrition group.

“In the aqua feed space, because it is such a narrow spec of what can be used — it needs to be concentrated, it needs to be low in fiber, there are specific amino acids that must be included — the alternatives are really quite few,” Wathne said.

That’s where the new canola seed comes in. Oil extracted from canola can be a 100 percent replacement for fish oil in its feed formulation, DeBonte said. This past fall, the company completed its first feeding study in Chile and found the fish grew at the same rate, had the same health and the same mortality rate as fish on a traditional fish oil diet.

“We relieve the pressure of a lot of the (small fish) harvest, and we aren’t creating new acres to grow our omega-3 canola. We have integrated it into our existing land,” DeBonte said. Cargill, responding to concern about soil degradation, says it will use crop rotation with two cereal grains, then a canola rotation, followed by a pea rotation.

But traditional fish meal sourced from the ocean also provides protein. EWOS has been weaning its reliance on this marine life-based fish feed by using plant-based alternatives, like soy, in its feed.

“We have reduced fish meal because it is a sustainability issue. The ocean’s potential is now harvested,” Wathne said. Soy raises other sustainable issues, like rain-forest deforestation and concern over fair treatment and pay for the soybean farmers.

Cargill announced in November a partnership with California-based Calysta Inc., which produces a protein for fish through the process of fermentation. Cargill will convert its Memphis, Tenn., plant, previously used for corn oil and sweeteners, to a create this product, called FeedKind. The process adds methane, oxygen and nitrogen to a fermentation tank where microbes then consume these gases. The process creates clean water, which is recycled back into the fermentation process while the microbes are converted to FeedKind protein. Carbon dioxide is the only byproduct.

“Now we have a novel protein that is almost a copy of fish meal in composition,” Wathne said. “It has very few limitations for how big it can become.”

The Memphis plant will eventually produce 200,000 tons of feed a year. Wathne said it takes 1 million tons of fish to produce that amount of feed. “That’s in the area of 25 percent of the annual Peruvian quota, which is the biggest producer. So you are actually saving 1 million tons of fish that can be used for other purposes.”

The two projects are separate, hold different strengths and weaknesses, but address two critical ingredient problems. FeedKind is non-GMO, non-plant, non-animal, non-water based while the canola seed is genetically modified. But FeedKind doesn’t currently contain omega-3s like the canola plant, which is one of the most critical ingredient needs for human health benefits. And the canola plant is grown in nature rather than in a factory, which could affect public perception.

Cargill still must gain regulatory approvals from the U.S. government for both products. FeedKind has been approved for use in the European Union. The company hopes to begin selling the omega-3 enriched canola product in 2020 while Cargill’s Memphis plant should be churning out FeedKind in late 2018.