NERC Issues Level 3 Alert As Grid Faces `Unprecedented Challenges' Due to Surge In Large Power Consumers

#19,158

Twenty years ago (2006) NERC, or the North American Electric Reliability Corporation, was tasked with "ensuring the reliability of the North American bulk power system" following the 2003 Northeast blackout which affected more than 50 million people in the United States and Ontario, Canada.

Over the past dozen years we've looked at a number of their reliability assessments (see here, here, and here) amid growing governmental concerns over the reliability of the electrical grid (see NIAC: Surviving A Catastrophic Power Outage).

Until relatively recently, the biggest threats to the grid were thought to be natural disasters (hurricanes, ice storms, severe space weather, etc.), `bad actors' (cyber-threats, sabotage, etc.), or aging infrastructure (see ASCE report card on America’s infrastructure).

But over the past couple of years the rapidly increasing power demands from A.I. data centers, bitcoin harvesting, and cloud computing have added yet another potential failure point.

Last summer the U.S. Department of Energy published a 73-page report that warned that if current schedules for retirement of reliable power generation (especially baseload) continue, without enough firm replacement, the risk of blackouts in 2030 could increase by 100× over current levels.

Four months ago, in NERC: Long-Term Reliability Assessment (Jan 2026), we looked at a 181-page NERC Long-Term Reliability Report which similarly warned that our power grid is facing a growing risk of electrical shortfalls over the next decade.

From their press release:

January 29, 2026

WASHINGTON, D.C. – NERC’s 2025 Long-Term Reliability Assessment (LTRA) and infographic spotlight intensifying resource adequacy risks throughout the North American bulk power system (BPS) over the next 10 years. Summer peak demand is forecast to grow by 224 GW, a more than 69% increase over the 2024 LTRA forecast with new data centers for artificial intelligence and the digital economy accounting for most of the projected increase.

Winter demand growth continues to outpace summer demand growth with 246 GW of growth forecast over the next 10 years, reflecting the evolution of electricity usage. Uncertainty and lag in the pace of new resource additions are driving heightened concerns that industry will not be able to keep up with rapidly increasing demand.

Up until recently, the biggest concern has been limited generating capacity along with increasing demand, but last September NERC issued a Level 2 Alert which warned of a new threat; that the power draw from these massive computing centers can be erratic, with sudden drop offs and surges, that can destabilize the grid. 

They wrote: 

NERC, Regional Entities, and NERC registered entities have analyzed a series of disturbances that occurred on the bulk power system (BPS) resulting in widespread and unexpected customer-initiated load reduction of large loads. These disturbances involved multiple events during which 1,000+ MW of unexpected Large Loads output reduction occurred, with most events occurring in 2024 or 2025. The increase of Large Loads-related events coincides with an increase in Large Load penetration across the BPS.

Since then, it has become apparent that:

the risks are increasing faster than expected,

real-world events are already occurring,

and industry response to the earlier alert has been insufficient

All of which forced NERC to raise the ante last week. 

NERC Issues Level 3 Alert, Reliability Guideline Focused on Large Load Challenges
May 04, 2026
 
WASHINGTON, D.C. – As the grid faces unprecedented challenges from a surge in large power consumers, NERC is taking significant steps to ensure the reliability of the bulk power system (BPS). NERC released a Level 3 Essential Action Alert, Computational Load Modeling, Studies, Instrumentation, Commissioning, Operations, Protection, and Control, outlining seven actions registered entities must implement to address immediate risks posed by computational loads interfacing with the BPS.

The Level 3 Alert was issued as NERC observed customer-initiated large load reductions and significant oscillations that occur in seconds, leaving little or no room for real-time responses, threatening BPS reliability. The deadline for registered entities to submit their responses is August 3, 2026.

In another move to address emerging large loads, NERC released new voluntary guidelines to safeguard grid reliability. The Reliability Guideline: Risk Mitigation for Emerging Large Loads, recommends actions for traditional utilities and grid operators, and the companies behind these large loads including equipment manufacturers. The goal is to ensure that as more industrial-scale consumers connect to the grid, they actively participate in practices that protect grid stability. These steps highlight that proactive planning and participation can enable even more of these facilities to come online reliably and quickly. The guideline also acts as a reliability bridge while NERC updates its formal Reliability Standards to address these new challenges.

Registered entities subject to the Level 3 Alert are encouraged to act now by reading the alert and submitting responses by the August deadline. And, although non-binding, NERC strongly urges all relevant entities, from transmission operators to equipment makers, to adopt the recommended risk mitigation strategies outlined in the Reliability Guideline.

 Those interested in reading the full 15-page document can find it at: 

 
Although the risk mitigation strategies mentioned above are currently non-binding, it seems likely this is a prelude to a major policy shift, as they call them `. . . a reliability bridge while NERC updates its formal Reliability Standards to address these new challenges.'

While it remains to be seen how much of an impact these massive data centers will have on day-to-day delivery of electricity to the nation, it seems likely that utility costs will continue to increase, and that localized brownouts/blackouts will become more common. 

With hurricane season approaching, those interested in small solar power options to soften the impact of power outages may wish to revisit the following blogs.

The Gift of Preparedness - Winter 2023 Edition

 #NatlPrep: Prolonged Grid Down Preparedness

How Not To Swelter In Place

Preprint: Variable Transmission Efficiency of Mammalian Origin HPAI D1.1 H5N1 Strains in Ferrets

 

#19,157

When we talk about the dangers of the H5Nx virus, it is with the understanding that there are currently several genetically distinct subclades of the virus circulating around the globe (2.3.4.4b, 2.3.2.1a, 2.3.2.1e, etc.), spanning numerous subtypes (H5N1, H5N2, H5N5, H5N6, H5N8, etc.) and literally hundreds of genotypes.

Fortunately, these HPAI viruses are not all created equal. Most only infect birds, while some have evolved enough to spillover into mammals, while fewer still have shown the ability to infect humans. 

Human infections may be mild - as we've typically seen with the North American  `Bovine' B.13 genotype of H5N1 (clade 2.3.4.4b) -  or more severe - as we've seen with the clade 2.3.2.1e H5N1 viruses (40% CFR) found in Cambodia.

Even when you compare nearly identical viruses (same subclade, same subtype, and same genotype) some have proved more virulent, or more transmissible, than its close relatives. 

Which is why we often find ourselves focusing on tiny amino acid changes in the influenza genome - including PB2 mutations like E627K, D701N, Q591K, and M631L and HA mutations like Q226L and E190D - which may favor mammalian adaptation. 

But new ones (see Sci. Adv.: PB2 and NP of North American H5N1 Virus Drive Immune Cell Replication and Systemic Infections) continue to be discovered.

The impact of these mutations can vary between virus strains, and can be enhanced or attenuated by other amino acid changes (some known, others yet to be discovered) that may occur elsewhere in the genome. 

For years, researchers have looked at the PB2-E627K mutation as possibly being the most important, but we've a preprint today that finds - at least in one specific H5N1 D1.1 strain -  that the PB2 D701N mutation appears to drive virus transmissibility in ferrets. 

First the link, abstract, and the discussion from the preprint.  I'll have a bit more after the break. 

Variable transmission efficiency of mammalian origin HPAI D1.1 H5N1 strains in ferrets

Grace E. Quirk,  Michelle N. Vu,  Valerie Le Sage,  Kaitlyn Bushfield-Thomason,  Hanh Dung Nguyen,  Seema S. Lakdawala
doi: https://doi.org/10.64898/2026.05.07.722809
This article is a preprint and has not been certified by peer review 
 
Preview PDF

Abstract

Highly pathogenic avian influenza H5N1 2.3.4.4b genotype D1.1 lineage continues to predominate in the United States wild bird population and has spilled over into dairy cattle three independent times. To assess the transmission risk of this sublineage, we performed direct-contact transmission experiments for three distinct D1.1 strains in ferrets. 

Two of these strains were isolated from humans and one from a lethal cat infection. We found that only one human isolate (A/NV/10/2025) was able to transmit efficiently between ferrets. Compared to the other strains, this isolate harbored the mammalian adaptive PB2 D701N mutation, suggesting this mutation may be critical for D1.1 transmission as opposed to the PB2 E627K substitution present in the lethal cat isolate. 

Based on these data we conclude that the transmission fitness of D1.1 strains is modest but that special attention should be paid to emergence of adaptation at the PB2 701 position.

(SNIP) 

Discussion

In this study we describe transmission fitness of three D1.1 genotype 2.3.4.4b clade H5N196 viruses. Two of the three strains, A/WA/255/2024 and A/Cat/TX/009022-007/2025, had low to no spread to cohoused ferrets. Reports by the CDC using a related D1.1 strain A/WA/239/202498 reported low transmission efficiency by direct contact to 1/3 recipient ferrets [9].

In contrast, A/NV/10/2025 spread to all three cohoused animals. Additionally, recipient ferrets naturally infected with A/NV/10/2025 reached humane endpoint criteria prior to the scheduled end of the study, suggesting that not only is this virus transmissible, but it can result in severe natural infection. 

The viruses in this study were chosen because of their mammalian origin and their unique PB2 mammalian adaptive residues. A/NV/10/2025 has a classical PB2 mammalian adaptive mutation at position 701, in contrast to the 627 residue present in the A/Cat/TX/009022-007/2025106 strain.

While we cannot exclude the impact of other mutations in present in A/NV/10/2025 (Table107S2) compared to A/Cat/TX/009022-007/2025 and A/WA/255/2024, it is highly likely that mammalian adaptation of residue 701 is playing a role in the forward transmission potential of this strain. 

Interestingly, only 1% of all D1.1 strains curated by Nextstrain [10] contain a PB2110 D701N mutation, but 7% of spillovers into mammalian species contain this mutation [3]. While PB2 E627K mutation is more prevalent in mammalian spillover viruses at 35%, this mammalian adaptation in D1.1 has not been related to forward transmission of the virus in ferrets in our study or others [9]. 

Our findings suggest that while less common in initial mammalian spillover events, PB2 D701N may increase the risk of onward transmission. Targeted surveillance of D1.1 strains for the acquisition of PB2 D701N in animals at the human-animal interface may identify priority variants for pandemic risk assessment. 

       (Continue . . .)

We've looked at the impact of the PB2-D701N mutation often in the past, and while considered to be an important mammalian-adaptation marker in both H5 and H7 subtypes, it has not been a reliable predictor of transmissibility or virulence across the board. 

But for the H5N1 D1.1 virus (and perhaps others), the presence of the PB2-D701N mutation may be a useful early warning sign. 

I would note that just two days ago, in Sci Adv: Mammary and Respiratory Infection of Sheep with H5Nx clade 2.3.4.4b Viruses with Milk-mediated Transmission to Lambs, that one of the mutations seen in H5N1 infected sheep was PB2-D701N. 

Deep sequencing of milk samples from mammary glands, oral swabs, and lung tissues revealed the emergence of viral variants distinct from the consensus sequence generated from the challenge D1.1 virus.

Notably, the left mammary gland of a D1.1-infected sheep appeared to select for the variant PB2-701N, while the right gland retained mixed residues of PB2-627 (E/K) and PB2-701 (D/N). 

Another indication that while the emergence of a pandemic strain of H5N1 may be a long shot, Nature's laboratory is open 24/7, and it continues to tinker. 

At Least One Evacuated American Passenger From m/v Hondius Tests Positive for Hantavirus

 

#19,156

Overnight the HHS announced that - of the 17 evacuated Americans from the m/v Hondius - 1 (currently asymptomatic) individual has tested positive for Hantavirus, and a second individual has shown `mild symptoms'.  

While the mantra has been that testing of asymptomatic individuals was unlikely to yield positive results, this is a reminder that there are very few absolutes when it comes to viruses and human physiology. 

We discussed the possibility of asymptomatic spread of the Andes Virus a week ago, and while evidence is sparse, studies have suggested (see Serological Evidence of Hantavirus Infection in Apparently Healthy People from Rural and Slum Communities in Southern Chile) that at least some infections may be mild or asymptomatic.

A more recent 2025 study (see Virological characterization of a new isolated strain of Andes virus . . .), published in PloS NTD reported:

In this work, we described the isolation of the strain responsible for the largest ANDV PTP transmission outbreak, which occurred in the small town of Epuyén and began on November 2, 2018. This strain, ARG-Epuyén, exhibited a high capacity for PTP transmission, necessitating the implementation of quarantine measures to curtail further spread [8].

The median reproductive number (the mean number of secondary cases caused by an infected person) was 2.12 before control measures were implemented and subsequently dropped to below 1.0 by late January.

Early intervention allowed for the collection of samples leading to the isolation of this new ANDV strain from an asymptomatic case. An early passage of this strain was sequenced, revealing only one amino acid difference from the virus recovered from the patient. Like the Andes/ARG strain, this strain was able to grow in a new host without needing adaptation [26].

So while rare, there is precedent for positive test results from asymptomatic individuals.  

What isn't well understood is whether - or how effectively - asymptomatic (or presymptomatic) individuals may be able to transmit the virus. The ECDC's Threat Assessment Brief, published on May 6th, had this to say:

Do asymptomatic individuals have a role in transmission?

Current very limited evidence does not support a significant role for asymptomatic individuals in hantavirus transmission, supporting active symptom monitoring of asymptomatic exposed individuals. Infectivity is highest on the first day of symptom onset, which indicates a high likelihood of some infectiousness one-two days before onset of symptoms.

 
While a lot of governments, eager to reassure the public, are quick to equate asymptomatic with `healthy', the reality is far more nuanced. Given its lengthy incubation period, an exposed individual's status can change in a matter of hours. 

Which is why - while not wanting to use the dreaded `Q' word - exposed individuals around the world are being segregated and monitored for symptoms. 

While I remain far from convinced that this Andes virus outbreak will turn into a global public health emergency, there are enough unknowns here to command our respect and attention. 

And even if the current strain of ANDV is incapable of bigger things, evolution is a thing. 

Meaning that anything we say today about the virus may not hold true tomorrow. 

WHO WPRO: 1 (fatal) Human Infection with H5N6 Reported By China

 

#19,155

After an impressive run of cases between 2021-2023 (see ECDC chart below), we've gone nearly 2 years (July 2024) since the last human H5N6 case was reported by of China.

While it is certainly possible that there have been cases that were either not detected by local surveillance - or were simply not reported - we've continued to see studies coming out of China cautioning on the the evolution of this subtype.

Emerg. Microbes & Inf: Unique Phenomenon of H5 HPAI Virus in China: Co-circulation of Clade 2.3.4.4b H5N1 and H5N6 results in diversity of H5 Virus

Transboundary & Emerging Dis.: The H5N6 Virus Containing Internal Genes From H9N2 Exhibits Enhanced Pathogenicity and Transmissibility

China CDC Weekly: Infection Tracing and Virus Genomic Analysis of Two Cases of Human Infection with Avian Influenza A(H5N6) — Fujian Province, China

So it is not completely surprising that the WHO reported, in their most recent  Avian Influenza Weekly Update # 1044 SitRep, on China's 93rd human infection (since 2014) with H5N6.  

Human infection with avian influenza A(H5N6) virus

From 1 to 7 May 2026, one new case of human infection with avian influenza A(H5N6) virus was reported to WHO in the Western Pacific Region. The case is a 55-year-old female from Chongqing Municipality,China, with symptom onset on 16 April 2026. She developed severe pneumonia, was hospitalised on 23 April, and died on 3 May. 

She had purchased, slaughtered, and consumed poultry. Samples collected from a cutting board tested positive for influenza A (H5). All close contacts tested negative and developed no symptoms. Since 2014, a total of 93 laboratory-confirmed cases of human infection with influenza A(H5N6) virus including 58 deaths (CFR 62.4%) have been reported to WHO in the Western Pacific Region.

Once again, this case appears to be linked to the purchase of live market poultry.

As we've discussed previously (see Mixed Messaging On HPAI Food Safety), there is some degree of risk in the slaughtering of live birds and preparation of raw poultry; especially from birds raised at home or purchased from live markets.

In 2024 the WHO published  Interim Guidance to Reduce the Risk of Infection in People Exposed to Avian Influenza Viruses, which lists a number of `risk factors', including:

• keep live poultry in their backyards or homes, or who purchase live birds at markets;

• slaughter, de-feather and/or butcher poultry or other animals at home;

• handle and prepare raw poultry for further cooking and consumption;

Although far more common in Asia and the Middle East, dozens of outbreaks of HPAI H5 in poultry markets here in the United States have been reported (see USDA Report 9 More Live Bird Markets Infected With HPAI H5).

While reports of human H5N6 infection in China have receded sharply over the past couple of years, novel influenza A viruses have a nasty habit of reinventing themselves (often via reassortment), before making dramatic returns.

Which is why we never like to say `never' when it comes to novel flu.  

Sci Adv: Mammary and Respiratory Infection of Sheep with H5Nx clade 2.3.4.4b Viruses with Milk-mediated Transmission to Lambs

 

#19,154

While we await further news on the offloading of hantavirus exposed passengers from the m/v Hondius, we still have plenty of older and emerging threats to keep our eye on.  Late last week a new study was published in Science Advances by Canadian researchers who investigated the susceptibility of sheep to both HPAI H5N1 and HPAI H5N5. 

Since the first outbreak of HPAI H5N1 was reported in U.S. dairy cattle a little over 2 years ago, surveillance and testing of mammalian livestock has focused almost exclusively on lactating dairy cows.

Yet during this time, we've also seen sporadic HPAI spillovers into goats, alpacas, pigs, and sheep in the UK and the discovery of  H5N1 antibodies in sheep in Norway. Other non-bovine findings include:

Transboundary & Emerg Inf: Serological Evidence of HPAI (H5N1) in Invasive Wild Pigs in Western Canada,

Preprint: Highly Pathogenic Avian Influenza H5 Virus Exposure in Goats and Sheep (in Pakistan).

EID Journal: Evidence of Influenza A(H5N1) Spillover Infections in Horses, Mongolia

Livestock surveillance - even of cattle - remains passive and limited, despite the call from animal health authorities (see WOAH Statement (Oct 22nd): High Pathogenicity Avian Influenza (HPAI) in Cattle) to increase vigilance. 

Although cattle-centric, WOAH does refer to `cattle and other livestock populations' in their messaging, but until now spillovers into sheep, goats, and other non-bovie ruminants have been often dismissed as either rare or dead end infections.

Today's study illustrates that small ruminants are not only susceptible to HPAI H5 infections; there is the potential for `widespread virus transmission within flocks'.  

The authors discuss the need for increased surveillance and the implementation of stricter biosecurity measures, particularly in mixed-species environments.

First the link, abstract, and some excerpts from a much longer study.  Follow the link to read it in its entirety.  I'll have a bit more after the break.  

 

Mammary and respiratory infection of sheep with H5Nx clade 2.3.4.4b viruses with milk-mediated transmission to lambs
Tamiru N. Alkie , Carissa Embury-Hyatt, Anthony V. Signore , Estella Moffat , Darwin Ramos, Sugandha Raj, Tamiko Hisanaga, Ifeoluwa Ayilara , Daniel S. Sullivan, [...] , and Yohannes Berhane 9 authors Authors Info & Affiliations
Science Advances
8 May 2026
Vol 12, Issue 19

Abstract

H5Nx clade 2.3.4.4b viruses are evolving rapidly, expanding host ranges and threatening animal and public health. In the US, genotype B3.13 dominates dairy outbreaks, while D1.1 is linked to fewer cases. In the UK, an asymptomatic ewe infected with genotype DI.2 raised concerns about ruminant susceptibility.

We inoculated lactating and nonlactating sheep with D1.1 (H5N1) and A6 (H5N5) viruses. Intramammary inoculation in lactating sheep caused clinical mastitis, high viral loads in milk, and transmission to suckling lambs, which further spread infection to the uninoculated mammary glands. Both ewes and their lambs seroconverted.

Aerosol exposure of nonlactating sheep led to transient respiratory infection, with low-level viral replication, and seroconversion. In vitro, both viruses replicated in sheep mammary epithelial cells.

These findings establish sheep as a viable ruminant model for H5N1 and H5N5 infection and highlight previously unidentified transmission dynamics, including milk-mediated and lamb-to-ewe spread, relevant for surveillance and biosecurity in ruminant populations.

        (SNIP)

Deep sequencing of milk samples from mammary glands, oral swabs, and lung tissues revealed the emergence of viral variants distinct from the consensus sequence generated from the challenge D1.1 virus.

Notably, the left mammary gland of a D1.1-infected sheep appeared to select for the variant PB2-701N, while the right gland retained mixed residues of PB2-627 (E/K) and PB2-701 (D/N). Sequence analysis suggests that the PB2-D701N mutation arose before transmission to the left gland, as oral swab from one suckling lamb already carried this mutation.

Dairy cows experimentally infected with H5N1 via the intramammary route acquired PB2-E627K mutation (32). While D1.1 viruses from some dairy cases exhibited the mammalian adaptive mutation PB2-D701N, the genotype B3.13 from all dairy cases retained the bovine-specific PB2-M631L mutation (74). Both PB2-E627K and PB2-D701N provide IAVs with significant replication advantages in mammalian hosts and enhanced viral transmission (75, 76).

(SNIP)

The findings from this study have significant implications, even though there are limitations such as small sample sizes and the lack of assessment of transmission from inoculated mammary glands to uninoculated glands from the environment.

However, all experimental studies in dairy cows revealed the restriction of infections and virus replication to infected quarters only. Our experimental approach has demonstrated that small ruminants are susceptible to H5N1 infections. 

The detection of seropositive goats and sheep during periods of heightened H5N1 activity underscores the necessity for more extensive investigations within the small ruminant herds. Once mammary infections with H5N1 have occurred in some lactating ruminants, virus can spread between the udders of lactating ruminants during suckling as some neonates could access milk from cohoused multiple lactating mothers.

This scenario suggests potential for widespread virus transmission within the flocks. We suggest increased surveillance and the implementation of biosecurity measures, especially in mixed-species livestock systems or where large numbers of lactating ruminants and their neonates were cohoused or allowed to graze on communal pastures. 

Moreover, similar to infected dairy cows, milk obtained from infected lactating small ruminants was found to harbor higher levels of infectious viruses. This raises important zoonotic considerations in areas where raw milk consumption is common. Furthermore, the handling of infected small ruminants poses risks to human health, highlighting the need for thorough risk assessments to be carried out.

       (Continue . . . )

Whether sheep, goats, or other small ruminants will ultimately increase the risk of HPAI remains unknowable, but it does give the virus more places where it can hide, thrive, and potentially make genetic improvements. 

Mixed-species farms are particularly worrisome, as they provide novel viruses with access to numerous hosts, and fresh opportunities to reassort or adapt (see Study: Seroconversion of a Swine Herd in a Free-Range Rural Multi-Species Farm against HPAI H5N1 2.3.4.4b Clade Virus).

While the evidence of HPAI H5's growing host range continues to mount, the $64 question remains; can we adapt to the threat faster than the threat is adapting to us?

Stay tuned. 

ECDC: Rapid Scientific Advice on Management of Hantavirus Exposed Passengers

#19,153

The ECDC has released their own set of recommendations for EU countries regarding the management of passengers returning from the m/v Hondius.  It should be noted that this is an advisory document, and not a mandate.

While these types of documents carry considerable weight, each member state can decide how to best manage the situation.

I've provided the the summary, and a link to the full 8-page PDF below.

Rapid Scientific Advice on the management of passengers - In the context of the Andes virus outbreak on the cruise ship MV Hondius

Assessment
9 May 2026

As of 9 May 2026, a total of eight cases of Andes virus infection, including three deaths and one critically ill patient, linked to the M/V Hondius cruise ship have been reported.

Key messages

ECDC has classified all people on board the ship and for the purpose of disembarkation and repatriation to be high-risk contacts.

• Monitoring/quarantine up to six weeks (42 days); Day 0 = 6 May 2026.
• High-risk contacts: self-quarantine, daily symptom monitoring, test if symptomatic.
• Low-risk contacts: passive monitoring; isolate and test if symptoms develop.
• Flights: trace contacts for probable/confirmed cases only (same row ±2 rows on long flights).
• IPC: masking, one to two metres distancing, PPE for healthcare/cleaning
• Strong risk communication and misinformation management.

This document provides advice for public health professionals in the EU/EEA managing individuals potentially exposed to ANDV, including on:Defining contact classification criteria based on level of exposure, including close and prolonged contact with symptomatic people;

• The identification, management and monitoring of contacts, including advice on testing.
• Appropriate infection prevention and control (IPC) measures for managing repatriated passengers and crew, suspected and confirmed cases and their contacts in healthcare and community settings; and
• Risk communication, community engagement and the management of misinformation.

ECDC rapid scientific advice disclosure statement: ECDC issues rapid scientific advice to meet an emergent or urgent public health need or to quickly reply to external requests. To accommodate the accelerated timeline, the process and methods used for the development of rapid scientific advice may be modified from those of standard assessments and recommendations. Potential limitations are described.

Publication file

Rapid Scientific Advice on the management of passengers - In the context of the Andes virus outbreak on the cruise ship MV Hondius