I was flying from the Woods Hole Oceanographic Institution, where I work, to Japan, the site of the Fukushima Daiichi nuclear power plant.

The plant was devastated by a tsunami on March 11, 2011. It was reportedly releasing dangerous levels of radioactive materials, most of which was ending up in the ocean. My job, along with a handful of other marine scientists, was to survey the contamination in the surrounding oceans. My expertise as a marine radiochemist would come in handy. I was worried about what we might find.

I’ll never forget what it felt like when I stepped out of the taxi near the coast that had been hit by a 9.0 earthquake and 15 meter-high waves only weeks before.

The damage went on and on: All I could see, for miles, was flattened earth. Almost every building, tree or structure that had once been there was gone, reduced to soggy rubble. Piles of cars, debris from houses and vegetation dotted the landscape, all awaiting disposal. I was standing on a spot where more than 18,000 people died or went missing; the thought was staggering.

But I was there to work.

Soon after arrival, I boarded a research boat sent from Hawaii to measure radioactivity—both types and amounts—in the nearby ocean.

We were always at least 30 kilometers from the shoreline, but even that far away the ocean contained debris dragged out by monstrous waves, around which our captain had to deftly navigate. Occasionally, we saw tree limbs, boxes and trash floating in the water, vestiges of a once-normal life that had been washed away.

The radioactivity levels in the ocean immediately following the accident were unprecedented—millions of times higher than what was there before. But heroic actions on land soon reduced the flow of contaminated water to the sea, and the ocean recovered quickly: By the time we got there in June, radioactivity close to the power plant was already about 1,000 times lower than at its peak in early April.

Still, the fish contained relatively high levels of cesium-137 and cesium-134, two products of nuclear fission, making them potentially unsafe to eat. Japan shut down the local fisheries and kept them closed for years. More than 100,000 fish have been tested since 2011, and since 2015, only a couple fish exceeded Japan’s strict limits for cesium. I have no hesitation eating seafood when I am in the region.

I’ve been back to the region roughly once a year since 2011.

It’s been encouraging to watch the marine life rebound without the pressure of local fisheries. Although we still use nets to gather plankton and other microorganisms for testing, a simple pole will often do to catch larger fish; these, I’ve been told, are now more prevalent in the nets of the local fishermen.

The shoreline looks nothing like it did before the accident. Spaces that once held housing and communities are still mostly open land, but the coastline is largely covered by bus-sized blocks of concrete meant to serve as barriers to prevent damage from any future tsunamis.

But not all aspects of the recovery are proceeding apace.

The Japanese government said in the past that everything is “under control,” but measurements from the ocean show that the reactors are still leaking radiation.

But these small leaks—which pose little risk to swimmers—aren’t what keep me up at night. Instead, 10 years after this devastating event, I and other experts are worried about the safety risks posed by the 1,000 tanks that together contain more than 1 million tons of radioactive waters, sitting at the power plant only steps from the shoreline.

This water grows in volume by roughly 100 tons each day, as groundwater still enters the buildings and mixes with the contaminated water used for cooling the damaged reactors. The future of these tanks needs to be decided.

We knew the tanks contain high levels of tritium, a radioactive form of hydrogen that is hard to remove from water during remediation efforts because water itself contains hydrogen as well. Luckily, at low doses, tritium causes less damage to living cells than many other forms of radioactivity.

In 2018, the Tokyo Electric Power Company (TEPCO), which operated the plant and is cleaning up the site, announced for the first time that the tanks also contain concerning levels of other, more harmful radioactive materials such as cobalt-60 and strontium-90, which are much more likely to end up on the seafloor or be incorporated into sea life.

We shouldn’t be hearing about this for the first time seven years after the accident. We should be getting more complete and accessible information. Although TEPCO regularly communicates with the public, the only data we have about non-tritium elements come from a fraction of the tanks—about 200—and don’t include other potential contaminants, such as plutonium.

The figures are also often buried in hard-to-find PDF files. To analyze these data, I have to type hundreds of numbers by hand into an Excel spreadsheet.

To win back the trust of the public and experts like myself, TEPCO and the Japanese government must do a better job of releasing data about the state of the remaining 1,000 tanks and demonstrate that they have cleaned up the non-tritium contaminants before they propose to release the water into the ocean. Independent assessments and monitoring of the ocean are needed.

We may need to give more consideration to ocean dumping alternatives, like continued and safer storage on land, until the radioactivity can naturally decay.

There is no time to waste: In February 2021, an earthquake near the site caused overflow of some of this deadly wastewater and prompted dozens of tanks to shift in their positions, though no evidence of ocean releases were reported.

Ten years after the nuclear disaster at Fukushima, we are still asking: Is it safe? Only with more transparency, better communication and continued independent studies will we begin to put this disaster behind us.

Ken Buesseler is a senior scientist at the Woods Hole Oceanographic Institution who studies radioactivity in oceans. The article was produced with Knowable Magazine.

The views expressed in this article are the writer’s own.