Premiering as part of Nature’s Season 44 on Wednesday, October 22, 2025 at 8/7c (PBS—check local listings), with streaming on pbs.org/nature, YouTube, and the PBS App, Walrus: Life on Thin Ice follows Smithsonian National Museum of Natural History Sant Director Kirk Johnson to Alaska’s Arctic, where roughly 250,000 Pacific walruses gather between Alaska and Russia. Johnson tracks females nursing calves on drifting sea ice near Nome, an orphaned pup’s rescue at the Alaska SeaLife Center, and males resting at Round Island—revealing how shrinking sea ice is reshaping behavior, crowding beaches, and raising survival risks. In the interview below, Johnson builds on these field moments: he reflects on what we can and can’t infer about population trends, the hazards of large haul-outs, gray-whale mortality signals, and what paleontology suggests about walrus resilience in a rapidly warming Arctic.

From your vantage point, what are the clearest lines of evidence that Pacific walrus numbers and distribution around Alaska are changing, and what’s driving those shifts most—ice loss, prey changes, or human disturbance?
I’m not an expert on this topic, and walrus counts are difficult because of the vast range and the fact that the population spans U.S. and Russian waters. Bud Fay’s 1982 walrus monograph showed how hard it is to census a population of marine mammals associated with sea ice, so the estimate of ~250,000 Pacific walruses likely had large error bars.

Improving techniques using satellite imagery and population genetics show great promise. The problem is that we don’t know whether Fay’s numbers were an undercount.

What we do see is the dramatic decrease of summer sea ice in the Arctic Ocean since satellite monitoring began in 1979 and the emergence of unusually large walrus haul-outs in both Russia and Alaska (e.g., Point Lay).
https://www.arcticwwf.org/the-circle/stories/pacific-walruses-in-russia/
https://arctic.noaa.gov/report-card/report-card-2015/walruses-in-a-time-of-climate-change/

Walruses rely on sea ice as a resting and nursing platform. What specific thresholds of ice extent or timing (spring/fall) are most critical for survival and reproduction?
I don’t think we know the answer to that yet. It does seem that crushing of juvenile walruses in large beach haul-outs and the exhaustion of nearby food sources contribute to mortality and undernourishment, respectively.

We’ve seen massive land haul-outs (e.g., Point Lay) when ice retreats. What do we know about stampede risk, calf mortality, and disease dynamics at these crowded sites?
What I’ve seen is largely anecdotal and based on the number of carcasses on the beach after the haul-out disperses.

How is the benthic food web (clams, worms, snails) responding to a warmer, more stratified Arctic, and what does that mean for walrus foraging efficiency and energy budgets?
Benthic-feeding marine mammals (walruses and gray whales) tend to forage in shallow water (<200 ft). I’m not aware of extensive benthic sampling that documents a decrease in productivity (but I’m not up on all of that literature).

The Unusual Mortality Events (UMEs) documented for northbound gray whales in spring suggest they are not acquiring adequate calories from their traditional feeding grounds in the northern Bering Sea and the southern Chukchi Sea.

From a paleontologist’s perspective, what does the fossil record tell us about how walruses and other ice-dependent megafauna responded to past warm periods? Are there analogs—or is today’s rate of change unprecedented?
A key point is the difference in depth between the shallow seas (northern Bering and southern Chukchi Seas) and the adjacent Pacific and Arctic Oceans. The seas are <500 ft deep; the oceans can exceed 8,000 ft. Because walruses and gray whales feed on the seafloor of these shallow seas, they are vulnerable to sea-level changes driven by ice ages.

The Ice Age (Quaternary) began around 2.5 million years ago. When continental ice sheets were at their maximum, sea level was ~350–450 ft lower. Areas that are now the shallow Bering and Chukchi Seas were exposed as land and vegetated with tundra or arctic grasslands. Glacials and interglacials alternated many times, meaning these extensive feeding grounds were available mainly during interglacials—walrus populations likely waxed and waned with sea level.

We also know the earliest tusked, suction-feeding walruses are known from 4–5-million-year-old deposits in southern California. Thus, animals similar to the modern walrus lived before the Ice Age and much farther south than walruses do today. Both facts suggest walruses can adjust to fluctuating feeding areas.

Conditions over the last 10,000 years have been optimal for walruses, and the pre-19th-century population (before whaling and “walrusing”) may have been peak walrus. For these reasons, I’m optimistic about how walruses will respond to a warming climate. I would expect the population to decrease but not face extinction. Other human factors—shipping, hunting, pollution, etc.—may also matter if the Arctic continues to lose summer sea ice.

The Smithsonian holds deep historical collections. How are museum specimens—teeth, bones, baleen, stomach contents, archived tissues—being used to reconstruct diet, contaminants, disease, and genetics across time?
We have a remarkable and large collection of marine mammals, including skeletons, frozen tissues, teeth, baleen, earwax, parasites, etc., that have been used to study these questions.

Promising results have come from whale baleen and layered whale earwax.

What behaviors or moments you’ve witnessed in the field best illustrate “life on thin ice,” and what concrete policy or conservation actions (shipping lanes, disturbance buffers, protected haul-outs, emissions cuts) would most help walruses now?
The Bering Sea is vast and variable. I did two research cruises—one in 1982 (NOAA Discoverer) and one in 1985 (Univ. Alaska Alpha Helix). In June 1985, we were trapped in floating sea ice for a few days.

Since that time, there has been a dramatic decrease in sea ice in all seasons (see: https://nsidc.org/sea-ice-today/analyses/arctic-sea-ice-sets-record-low-maximum-2025). Local sea-ice distribution can be quite variable, but my direct observations in 1982 and 1985 gave me a personal baseline.

I’ve only watched wild walruses in 1985 and 2024, so I can’t claim to be a walrus expert, nor have I observed long-term behavioral trends. The emergence of very large haul-outs, the documented decrease in sea-ice cover, and evidence of benthic productivity decline suggested by unusual gray whale mortality events are the basic data underpinning the show’s narrative.
https://www.fisheries.noaa.gov/national/marine-life-distress/2019-2023-eastern-north-pacific-gray-whale-ume-closed