American Scientist

Open the heavy locked metal entrance door to the portal building of the Seed Vault, and the length of the tunnel becomes visible. The surface layer of rock is loose, the result of freezing and thawing for millennia, so the beginning of the tunnel is encased in heavy steel tube ( left ). Beyond this shallow, unstable, and fragmented layer of rock and soil, the permafrost area begins and the rock is solid.

Visitors instantly observe that the Seed Vault is not a pristine, antiseptic laboratory. It was not designed as a research facility or a tourist attraction. It’s a pratical facility all about freezing seeds. Pipes overhead carry cold air to the farthest reaches of the facility where the seeds are stored. Past the steel-tube section, the tunnel transitions to rock, spray coated with a layer of white plastic-fiber-impregnated concrete. It’s a bright tunnel, but austere. Having a facility that virtually operates by itself, without exposure to too many human interventions or potential mistakes, that does its job even if the mechanical cooling goes on the blink—all these factors are key to having a Seed Vault that functions properly over the long term, is secure, and is economically sustainable.

There is a set of closed, locked double doors at the very end of the tunnel. Open these and you enter an awe-inspiring, cavernous room in the heart of the mountain, a huge and colder room with tall ceilings coated with millions of tiny glistening ice crystals (right ). I have always thought of this as the “cathedral room.” It lies at right angles to the tunnel and serves as a staging area for preparing boxes to go into the frozen vault room.

The three vault rooms lie on the far side of the cathedral room. Two lie to the left of the tunnel shaft, one to the right. The layout is purposeful. None of the vault rooms are situated in a direct line down the tunnel from the entrance. Instead, at the architect’s behest, the workers carved out a concave indentation in the rock as a security measure. Were there to be any explosion inside the facility, this feature would direct the shock waves back out the tunnel, diverting them from the vault rooms on either side.

The seed storage vault rooms are located both where the natural temperature was coldest and where the quality of overhead rock was the best for structural integrity. Each of the three vault rooms is approximately 27 meters long, 9.5 meters wide, and 5 meters high. Only one room is currently needed and in use ( left ). We might discover that seeds of some crops are best conserved at a different temperature than the standard –18 degrees Celsius provided, which would indicate the use of one of the other rooms cooled to a different temperature.

Seed collecting did not begin with the Seed Vault. In fact, the Vault has never been involved with gathering seeds directly from the field. In the 1970s agricultural scientists sounded the alarm that the wave of modern seed varieties sweeping across the world was massively displacing traditional varieties and driving them into extinction. Today there are 1,700 collections of crop diversity. In total, these genebanks house more than 7 million samples, up to 1.5 million of which are thought to be distinct.

In the early 1990s I oversaw the United Nation’s first comprehensive global assessment of the state of the world’s crop diversity. What my small team and I found was shocking. The “global system” of genebanks was no system at all. There were a handful of effective facilities. Even the best genebanks, however, were often succeeding solely because of the heroic action and dedication of their unsung staff. How long was the world willing to rely on such luck? If this was plan A, what was plan B? We needed one, fast. And this is a big reason why having a safe “backup copy” in a place like Svalbard is so critically important.

The Vault now contains, for example, a sizable collection from the U.S.-based Seed Savers Exchange, the largest organization of gardeners working to conserve heirloom vegetable varieties (its members are shown at an heirloom tomato tasting at top right, and some of its tissue culture conservation techniques for heirloom potatoes are shown second from top). Another resident sample, collected in Colombia, called carcha, belongs to the gene pool of lima beans, with ancient Andean origins, possibly 6,000 years ago ( left, bottom ). And another genebank, the International Rice Research Institute in the Philippines ( left ), has deposited more than 120,000 types of rice in Svalbard.

It is difficult to estimate how much crop diversity exists in the world today, and impossible to know how much has been lost. In the fields of many traditional farmers, there are few “uniform” varieties wherein all plants are essentially identical. A wheat field may contain a number of different types, maturing at different times, with different degrees of pest and disease resistance ( below ).

As fighting broke out in Tunisia and elsewhere in what is now called the Arab Spring, I got on the phone with my old friend Mahmoud Solh, the director general at the International Center for Agricultural Research in the Dry Areas (ICARDA), then based in Syria. Mahoud and I agreed that Syria, and the ICARDA collection, would likely escape the turmoil. (How wrong we were.) But I suggested to Mahmoud that we should try to get a duplicate copy of his collection up to Svalbard as soon as possible, just in case. ICARDA was able to get seeds of virtually everything up to Svalbard—116,000 different varieties, more than 58 million seeds. In early September 2015, ICARDA requested the return of seeds from its duplicate stores in Svalbard ( sample seeds shown at right ). On September 23, 128 boxes of seeds, samples of 38,073 varieties of wheat, barley, lentil, chickpea, faba bean, grass pea, and a number of samples of crop wild relatives and forages were sent to Morocco and Lebanon, where ICARDA is reestablishing working genebanks. This first withdrawal of seeds is bittersweet. Everyone hopes this will be the last time the Seed Vault will be used for the purpose for which it was established. But I know it won’t be.