Arctic temperatures are rising more than twice as rapidly as the rest of the world, with myriad impacts on ice sheets, sea ice, snow cover, and many other crucial aspects of Arctic ecosystems and communities (Richter-Menge et al., 2019). One impact of this warming is the widespread thawing of permafrost—the thick subsurface layer of soil that (historically) remains frozen throughout the year. Permafrost thaw has many implications of importance to human society such as exacerbating climate change through the release of stored carbon, and posing risks to infrastructure as the permafrost gives way. There is also a growing realization of the possibility of adverse impacts on human health. Many reports discuss the potential impacts of increasing temperatures on tropical infectious diseases becoming more prevalent as the habitable zones for disease vectors expands (e.g., USGCRP, 2016). In recent years, a few scientists have begun to raise questions about whether infectious disease risks could possibly emerge or re-emerge from higher latitudes, including from melting Arctic ice and thawing permafrost (NRC, 2014).
This interest grew when, in summer 2016 (a year that shattered records for Arctic-region warmth), a remote part of Siberia known as the Yamal Peninsula saw an outbreak of anthrax, which killed a 12-year-old boy, infected dozens of people, and killed more than 2,300 reindeer. While the combination of factors that led to this outbreak is debated, one theory is that it may have resulted from the thawing of a frozen reindeer carcass infected with anthrax, which released spores into nearby water and soil. It is well known that anthrax forms spores (Bacillus anthracis is the bacterium that causes anthrax) that can persist for decades and remain viable to cause disease, but this phenomenon may now be exacerbated by permafrost thawing.
In addition to this known, observed threat, there is growing speculation about other types of bacteria and viruses that could emerge from ice and permafrost. In 1997, researchers obtained tissue from corpses buried in permafrost in Alaska that yielded influenza RNA fragments that facilitated recreating the sequence of the 1918 influenza virus (Taubenberger et al., 2007). While that work did not retrieve live virus, a 2005 NASA study revived bacteria that had been encased in Alaskan permafrost dated to the Pleistocene epoch (~12,000-2.5 million years ago) (Pikuta et al., 2015). Another research team in 2007 took ice from Antarctic mountains, two samples that had a minimum age of 100,000 years and the other of 8 million years, and resuscitated multiple bacterial populations and postulated that lateral gene transfer could occur by extant organisms upon thawing (Bidle et al., 2017). A 2014 publication reported on two types of “giant viruses” trapped in Siberian permafrost for 30,000 years and revived in the lab (Legendre et al., 2014). Although, these viruses were infectious only to single-celled amoeba, this work did raise concerns about the survival and emergence of viruses that
are infectious to humans. Such concerns are bolstered by the facts that fragments of DNA and RNA from diseases such as smallpox, bubonic plague, as well as the 1918 influenza virus have been recovered from permafrost (Theves et al., 2011, 2017; Zhang, 2006), and that many areas around the Arctic have buried human and animal remains harboring such pathogens. Looking back further in time, permafrost may harbor infectious viruses or bacteria that have been dormant for thousands or even millions of years (Houwenhuyse et al., 2018), for which local populations lack immunity and no countermeasures exist.
Given the vast and sporadic geographic distribution of these events and the lack of a robust surveillance system to identify cases in these regions, only limited numbers of studies have been published thus far. It is therefore difficult to reliably characterize the magnitude and nature of these potential risks. They may in fact be quite small compared to known risks from existing infectious agents that are prevalent at lower latitudes, however, understanding and preparing for “low-probability, high-consequence” events is one of the hallmarks of a robust public health protection strategy.
The National Academies of Sciences, Engineering, and Medicine (National Academies) in collaboration with the InterAcademy Partnership (IAP) and the European Academies Science Advisory Committee (EASAC) held a workshop in November 20191 to bring together researchers and public health officials from different countries and across several relevant disciplines to explore what is known, and what critical knowledge gaps remain, regarding existing and possible future risks of harmful infectious agents emerging from thawing permafrost and melting ice in the Arctic region. In total, 56 leading researchers from 15 different countries including representatives from Arctic indigenous communities were in attendance. The workshop included 11 individual talks, interspersed with 6 panels and group guided discussions, as well as 2 breakout sessions to examine case studies such as the specific case of Arctic-region anthrax outbreaks, as a known, observed risk as well as other types of human and animal microbial health risks that have been discovered in snow, ice, or permafrost environments, or that could conceivably exist. The workshop primarily addressed two sources of emerging infectious diseases in the arctic: (1) new diseases likely to emerge in the Arctic as a result of climate change (such as vector-borne diseases) and (2) ancient and endemic diseases likely to emerge in the Arctic specifically as a result of permafrost thaw. Participants also considered key research that could advance knowledge including critical tools for improving observations, and surveillance to advance understanding of these risks, and to facilitate and implement effective early warning systems. Lessons learned from efforts to address emerging or re-emerging microbial threats elsewhere in the world were also discussed.
1 This Proceedings of a Workshop was prepared by a workshop rapporteur as a factual summary of what occurred at the workshop. The planning committee’s role was limited to planning and convening the workshop. See the appendixes for the Statement of Task, planning committee biosketches, workshop agenda, and participant list.
Following welcoming remarks from the planning committee Chair, Diana Wall, Colorado State University; Vice Chair, Volker Ter Meulen, InterAcademy Partnership; and Henrike Hartmann, Volkswagen Foundation, a series of three keynote speakers set the stage for workshop discussions. Dr. Vladimir Romanovsky, University of Alaska Fairbanks, shared context-setting remarks on the history and current dynamics of permafrost. Permafrost2 includes any earth material at or below zero degrees Celsius for two or more consecutive years. Figure 1 shows the distribution of permafrost in the northern hemisphere, mostly located in the high latitudes and high elevations. The “active layer” is the layer above permafrost that thaws and freezes on a yearly cycle. Permafrost typically ranges from very small thicknesses near zero in discontinuous
2 “Permafrost is a permanently frozen layer below the Earth’s surface. It consists of soil, gravel, and sand, usually bound together by ice.” Source: https://www.nationalgeographic.org/encyclopedia/permafrost/.
permafrost regions to more than 1,500 meters thick in parts of Siberia, and the active layer thickness typically ranges from 30 centimeters to 1.5 meters from the surface.
Permafrost is a product of cold climates and is therefore sensitive to climate change. Dr. Romanovsky noted that recent changes in climate and warming temperatures on a global scale, especially in the high latitudes, have led to increased average temperatures that can be 2-3 times larger in high latitudes than temperatures in low latitudes. Permafrost reacts to these changes and temperature increases, which in some cases, has been as high as 2-3 degrees Celsius over the past few decades. A significant event in the polar regions, permafrost thaw can potentially have severe impacts on local communities, infrastructure, and ecosystems (Richter-Menge et al., 2019; Schuur et al., 2015). Future permafrost changes can be more readily understood in the context of past permafrost change.
For the purposes of this workshop, Dr. Romanovsky pointed out that researchers are primarily interested in permafrost from the last interglacial cycle (Figure 2). Permafrost that is currently in danger of thawing includes: (1) permafrost from the previous glacial period (approximately 12,000 – 126,000 years ago), (2) permafrost that resulted from cooling after the climatic optimum about 10,000 years ago, or (3) very recently formed permafrost as a result of cooling during the Little Ice Age (14th through 19th centuries). After the maximum extent of permafrost ~20,000 - 18,000 years ago, the climate began to transition into a recent relatively warm period (the Holocene), when that ancient permafrost experienced degradation. Conditions in Eurasia were very different at that time and included a tree-less environment, with windy conditions, cold temperatures, and rapid deposition of dust material. Despite the cold temperatures, ample vegetation was available to sustain various types of animals. Permafrost from this period contains a great deal of organic material including the remnants of vegetation and buried animals. Understanding when and how this organic material thaws will be important.
By the Holocene optimum (~9,000 – 5,000 years ago), most permafrost disappeared from Europe because of warming temperatures, especially at the continental shelves due to sea level rise. Following that period, the climate cooled again, and for the past 5,000 years, new permafrost has formed, largely reflecting present-day distribution. During the Little Ice Age, new permafrost formed, usually near the southern boundary of permafrost extent. However, when permafrost degrades, the newest permafrost thaws first. After the conditions of the Little Ice Age ended, the most recently formed permafrost began to thaw or disappeared completely. Late Holocene permafrost was stable for some time, but it has also now started to thaw. This thaw is widespread and slow. Dr. Romanovsky noted that in some places, there is currently a layer of former permafrost that does not freeze at all during the winter, an important threshold identified by researchers. Gradually, parts of the upper permafrost are introduced to the active layer (which freezes in winter and thaws in summer), and are incorporated into this seasonal process. Although this transition of permafrost to the
active layer is only on the order of a few 10s of centimeters, it is a widespread occurrence and the volume of material involved could be substantial. The changes are not linear year to year, but over the long term, there is a steady increase in the freeze/thaw depth.
Dr. Romanovsky also remarked on the importance of understanding abrupt permafrost thaw. With this type of abrupt change, he noted that the age of the permafrost is not as important as the processes on the surface, which have a greater impact. Some climatic and weather conditions can cause permafrost to thaw deep enough to melt ice and create ground subsidence (i.e., sinking or settling of the ground surface), the ground subsidence then fills with water, which causes the process to accelerate on itself. Weather conditions, forest fires, and other disturbances can trigger these conditions. Permafrost degradation under lakes, movement of thawed materials
near riverbanks, and coastal erosion due to increasing temperatures may also contribute to abrupt thaw. These processes are local but the thaw persists and accelerates through all ages of permafrost. Questions remain regarding the significance of these two different processes (annual seasonal permafrost thaw versus abrupt events) for release of microbes or genetic material and which of these may be more likely to lead to emergence events.
Dr. Romanovsky concluded his remarks noting that ongoing modeling efforts project continued climatic warming. Even assuming relatively moderate warming, late Holocene permafrost and permafrost from the last glaciation will continue to thaw by mid-century. By the end of the century, about 60 percent of permafrost will be in the process of thawing. With higher climate warming scenarios, about 75 percent of the north slope of Alaska permafrost will be thawing by the end of the century (Figure 3).
Dr. Albert Osterhaus, University of Veterinary Medicine Hannover, discussed emerging infectious and zoonotic diseases in the context of environmental change. In past decades, there have been several viruses and zoonoses at the origin of major
human disease outbreaks. The concept of “one health”3 encompasses human and animal diseases as well as the environment and the interactions between them. Approximately 75 percent of new human infectious diseases actually come from animals (Gebreyes et al., 2014). Transmission of obligate zoonoses (such as Trichinella) requires animals and can be acquired from eating meat of infected animals. On the other hand, it is possible for humans to directly acquire facultative zoonoses (such as influenza) from animals, but they are far more likely to acquire it from another human. Historical zoonoses (like HIV) originated as a cross-over of a simian virus, but are now maintained completely by human-human transmission. Following smallpox eradication and the subsequent cessation of smallpox vaccinations, there have been cases of new animal poxvirus crossing species barriers and causing infections in humans. Because vaccinations are no longer needed against smallpox, there may be a possibility, with the thawing of permafrost, that some human or animal carcasses will pose a smallpox or other disease threat to humans, he concluded.
Dr. Osterhaus pointed out that higher ambient temperatures in the Arctic could result in increased foodborne diseases, waterborne infections in humans, and changes in the migratory pathways of animal host populations and their contacts including changing rodent and fox populations, as well as the northern range of vectorborne diseases (Parkinson and Butler, 2005). For example, a 2005 study showed that an outbreak of Vibro parahaemolyticus (which causes gastrointestinal illnesses in humans) occurred in farmed Alaskan oysters when the temperature of the seawater rose above 15 degrees Celsius (McLaughlin et al., 2005; Figure 4).
Other examples shared by Dr. Osterhaus include tick-borne encephalitis in Europe and Asia, primarily located in the “tick belt” that stretches across the continent. The width of the belt and the extent of viral disease spread by vector distribution is largely determined by the climate. When the climate warms, as is occurring today, the belt widens and extends disease spread into areas that have traditionally not been affected. West Nile Virus (transmitted to horses and humans by mosquitos and carried by birds), was first discovered in the U.S. in 1999 after being introduced by air travel of infected humans, and is spreading across the entire U.S., causing the death of a few hundred people every year. Importantly, the survival of the mosquitos that transmit this disease is dependent on the climate. As temperature warms, the distribution of these mosquitos has increased to parts of Canada and the Arctic (Figure 5). As noted earlier, animal populations are changing and modeling shows that this will likely continue, due in part to habitat change. As another example, some modeling studies show that Arctic fox populations are changing and that climate change will have an effect on cases of rabies in the Arctic (Huettmann et al., 2017).
3 “One Health is a collaborative, multisectoral, and transdisciplinary approach—working at the local, regional, national, and global levels—with the goal of achieving optimal health outcomes recognizing the interconnection between people, animals, plants, and their shared environment.” Source: https://www.cdc.gov/onehealth/index.html. See also Roger et al., 2016.
New molecular techniques and technologies enable the identification of viral pathogens more effectively than previous methods of isolating viruses. Many of the viruses detected in the past few decades have some connection to changing climate and changing conditions. Dr. Osterhaus discussed an example of the changing migratory patterns of harp seals and the associated spread of Morbilliviruses, which have been detected crossing species barriers (e.g., Jo et al., 2018, 2019). Full-length sequencing of viruses can be obtained much more quickly now than in the past, and combined with phylogenetic trees, the origins of a virus can be detected. This is a powerful tool to trace where viruses originate and how they cross species barriers (e.g., Jo et al., 2018, 2019).
Using new data and sequencing, analyses can be used to detect, for example, the origin and emergence of ancient hepatitis B viruses (Muhlemann et al., 2018). Material from corpses was used to detect fragments of DNA from thousands of years ago. The key unknown is how long the viruses can actually survive in the permafrost. It is known, for instance, that poxviruses and anthrax can survive for quite a long time, on the order of hundreds of years, but the possibility of finding whole genomes is unknown.
Migratory birds are infected with Influenza A viruses in a subclinical way, and can directly transmit viruses to other species including humans. In fact, it was not considered possible for birds to spread avian viruses to humans until the late 1990s when confirmed human cases were discovered in Hong Kong. Focusing specifically on H5N1, Dr. Osterhaus demonstrated that it has mutated and spread throughout South East Asia, Europe, and Africa. Although the disease is fatal in tufted ducks (and to some extent, pochards), it appears not to be fatal for some wild duck species, particularly mallards (Keawcharoen et al., 2008). These species are capable of surviving and spreading the disease, and their migratory patterns may potentially shift due to climate
change. This has implications for the spread of this disease beyond known areas (Verhagen et al., 2015). Most of the H5N1 outbreaks in European wild birds were concentrated in areas of open water, where surface temperatures were above freezing (Reperant et al., 2010).
Dr. Osterhaus also explored the factors that make a pandemic virus. An experiment of passing a virus over the upper respiratory tract of ferrets showed that only five mutations were needed to change from a virus that does not efficiently spread, into a virus that can be easily spread (Herfst et al., 2012; Linster et al., 2014; Munster et al., 2009; Russel et al., 2012). Although this type of lab experimentation can be controversial, a similar phenomenon was detected in nature where a virus (H10N7) spilled over from ducks to seals. The original virus was purely avian, but it changed over time to a virus that spread from seal to seal (Bodewes et al., 2016).
At the conclusion of his talk, Dr. Osterhaus noted that a mix of unprecedented and complex anthropogenic (or human-driven) changes are drivers of infectious disease emergence in animals and humans. This implies that several diseases that could not spread in the past can spread today. Potential reversal or mitigating options for these changes may be topics of interest to the scientific community, and preparedness could be based on increased capabilities in terms of syndromic surveillance in humans and animals, pathogen discovery, development of diagnostics, improved animal and mathematical modeling capacity, and platforms for pathogenesis study, preventive intervention, and therapeutics.
Dr. Keith Chaulk, Stantec shared views on increasing atmospheric and oceanic temperatures in the Arctic and the connections to the people living in the region. Many indigenous communities live in the Arctic (Figure 6), with the Arctic Council estimating approximately a half million people live in these regions.4 These Northern peoples are experiencing many rapid changes including:
- permafrost thaw, which negatively affects physical infrastructure;
- increasing rates of wildfires, which contribute to areas of poor soil and loss of critical habitat;
- changes to wildlife populations and animal distributions that may introduce new disease vectors; and
- altered freeze/thaw timing and changing ice patterns on both sea and land with effects on travel routes, safety, and access to local foods.
These types of environmental change can have negative effects on traditional land-based practices, cultural practices, and community health in general. Because northern areas lack significant agriculture, local food harvests of fish, caribou, and seals provide key sources of nutrients. However, there are issues associated with food security in the North related to poor-quality store-bought foods, limited choices, and high food prices.
Dr. Chaulk also covered the intersection of (and similarities between) traditional knowledge and science knowledge. He highlighted the key principles of traditional knowledge, as adopted by the Indigenous Peoples Secretariat5: traditional knowledge is (1) “a systematic way of thinking … applied to phenomena across biological, physical, cultural, and linguistic systems”; and (2) “a body of knowledge generated through cultural practices and lived experiences” … “developed and verified over millennia”. He emphasized that traditional knowledge can help further understanding of nature, through various forms of record keeping as well as a broad range of scientific undertakings that are diverse in terms of their scope, focus, and research practices. Both traditional knowledge and scientific understanding accumulate over time. Although both knowledge systems are imperfect, they are still reliable and repeatable and can be used to inform decision-making. They are also both self-correcting; traditional knowledge is highly adaptable and scientific understanding advances over time as paradigms shifts and models are improved and replaced. Knowledge transfer is another key element of both knowledge systems. Dr. Chaulk noted that teaching and learning, as well as oral communication, are fundamental to both systems. Both have experts, who are supported by third party endorsement through peer reviewers and community members.
5 The Indigenous Peoples' Secretariat assists in creating opportunities for Arctic Council Permanent Participants to present their causes, support the provision of necessary information and materials, and communicate information about their work. Source: https://www.arcticpeoples.com/.
Dr. Chaulk provided three examples of indigenous knowledge related to microbes. In the first example, he illustrated the process through which Southern Labrador Inuit place sealskins in bogs in the spring to allow microbes to remove the hair from the pelts. It is possible that changing temperatures could affect bog functioning. A second example demonstrates the importance of frost timing on the picking and gathering of Vaccinium vitis idaea (Lingonberry). Community elders do not allow this berry to be picked until after the first frost. The fruit ripens at this time and allows a fruit worm to emerge. Longer frost-free periods could have implications for this process. Finally, he discussed an infection called “seal finger” that is caused by bites and contact with seal products. It causes inflammation and can lead to an unusable finger, and the prevalence of this infection may be related to warming waters in the North. To improve scientific understanding of these examples and other emerging issues, there are many benefits to working with indigenous partners. It can help reduce program costs and
develop capacity for long term monitoring. Scientists may also be able to leave a legacy of training and education as well as gaining insights into ecosystem and microbial processes, Dr. Chaulk concluded.