Heatwave poses challenge to Japanese medical system already stressed by virus

Paul Somerville and Andrew Gissing, Risk Frontiers

In recent years, eastern Australia, like Japan, has experienced extremely high maximum temperatures that are consistent with patterns of global changes in climate. Fortunately, last summer’s heatwaves in Australia occurred before the prevalence of COVID-19, and if Australia is able to maintain its suppression of the virus, it may be able to avoid the compounding effects of those conditions. This briefing demonstrates that even with the low prevalence of the virus in Japan, these compounding effects can be significant.

The number of people showing signs of heatstroke or heat exhaustion has sharply increased recently. Temperatures soared to 41.1 C in Hamamatsu in central Japan on Monday (Mainichi Shimbun, 2020a), tying with the country’s highest-ever temperature, marked in Kumagaya near Tokyo in 2018.

The 2018 Heatwave

During the 2018 heatwave, Mainichi Shimbun (2018) showed that the 94 people who died included 26 fatalities in Tokyo, where the heat reached 40.8 degrees in the suburban city of Ome. Saitama Prefecture also reported nine deaths, while in the western part of the country, Osaka Prefecture had six, Mie and Hyogo five each, and Hiroshima saw four. Aichi Prefecture in central Japan also announced four deaths. (According to Slate (2020), more than a thousand people died from heat-related illnesses over the course of those few weeks).

When broken down by the gender of the victims, there were 52 women and 42 men (Mainichi Shimbun, 2018). All of them were 40 years old or older. Those in their 80s constituted the largest group with 37 deaths, followed by 22 in their 70s, 15 in their 60s, 10 in their 90s, five in their 50s and four in their 40s.

Among the victims, 28 fell ill while they were outside, and many were farming in their fields. As many as 36 people were found ill or unconscious while they were inside, due in several cases to broken air conditioners or electric fans. In some cities such as Yamato, elderly residents who live alone are monitored day and night by an elaborate system of motion sensors and communication protocols between city officials, residents and their relatives.

Older people tend to have difficulty recognizing when they are dehydrated. They face the risk of their conditions deteriorating before realizing it, even when they are not subject to searing heat. Lowering temperatures inside using air conditioning is important, but not all homes have air conditioners.

2020 Heatwave – Distinguishing heatwave symptoms from corona virus symptoms

On August 19, 2020, officials in Tokyo reported that 28 people died of heatstroke in the city during the eight-day period from August 12 to August 19, bringing the total number of fatalities in Tokyo in August to 131 (NHK, 2020). The Medical Examiner’s Office said that 11 of the 28 victims were in their 70s, ten were in their 80s, and about 80 percent of the victims were at least 70 years old. Eleven of the victims died at night and 27 died indoors, of whom 25 were not using air conditioners.

In the midst of this year’s heatwave, it is reported that medical workers worry that the similarity of heat stress symptoms to COVID-19 may place extra pressure on a health care system already creaking under the strain of the coronavirus pandemic (Mainichi Shimbun, 2020a).  There are times when medical personnel cannot immediately distinguish those suffering from heat-related illness from those with COVID-19 when the patient is feeling unwell with high fever because that is a symptom they have in common. Japan has a relatively small number of COVID-19 cases (Figure 1), with only 1,169 deaths so far.  The Japanese Health Ministry reported no evidence of excess deaths during April and May (the latest months for which data are available), and it is likely that undetected COVID-19 cases are contributing significantly to the numbers of heatwave deaths that are being reported.

Figure 1. COVID-19 cases in Japan.  Cases: 62,507; Deaths: 1,181; Recovered: 49.340.  Source: Worldometers (2020), 25 August 2020.

The problem posed by the pandemic is that treatment has to take account of the possibilities of both COVID-19 and heat-related conditions when staff cannot rule out the possibility of coronavirus infection. Amid reported public fears that mask-wearing to prevent the spread of the novel coronavirus could itself cause heatstroke or heat exhaustion, 12,804 people were taken to hospital across Japan between Aug. 10 and Aug. 16 for heat-related conditions, up from 6,664 people the previous week, according to the Fire and Disaster Management Agency.  There is a concern that this large number of patients being taken to the hospital may cause the hospital system to collapse if the heatwave continues.

Recent heatwave conditions in the United States have also seen authorities needing to adapt plans to account for the risks of COVID-19, with fears that people may be reluctant to leave their homes to seek cooler shelter due to infection risks. Adaptions have included restricting the number of people accommodated within cooling centres to allow social distancing.

Some resources complied by the Global Heat Health Information Network on COVID-19 and heatwaves are available here: www.ghhin.org/heat-and-covid-19.

Public Information on Heat Stress

The Ministry of the Environment is providing English-language information about the heat stress index on its website in a bid to prevent illnesses caused by intense heat, which has become a major threat to health and even life in Japan in recent summers (Mainichi Shimbun, 2020c).

The website, designed for viewing by both smartphones and personal computers, indicates the intensity of the heat effect throughout the country in five colors, from blue (almost safe) to red (danger). It also provides two-day predictions for the heat stress index, as well as data for each observation point nationwide.

The heat stress index, also called the Wet Bulb Globe Temperature (WBGT), is one of the empirical indices showing the heat stress an individual is exposed to. It is calculated incorporating factors such as humidity, sunlight and reflection intensities and atmospheric temperature.

According to the ministry website, the number of people suffering from heatstroke shoots up rapidly when the WBGT, which is denoted in degrees but is different from normal air temperature, exceeds the upper threshold of the “Warning” level (25-28 degrees), when the air temperature is between 28 and 31 degrees Celsius.

For the warning level indicated in yellow, people are advised to rest often. When the index is at the “Severe Warning” level of orange, people are advised to refrain from heavy exercise. At the “Danger” level shown in red, people should stop all exercise.

Figure 2. Screen capture showing the Ministry of Environment website providing heat stress index information. Mainichi Shimbun (2020d).



Mainichi Shimbun:

  1. https://mainichi.jp/english/articles/20180724/p2a/00m/0na/002000c

2020a. https://mainichi.jp/english/articles/20200822/p2g/00m/0na/039000c

2020b. https://mainichi.jp/english/articles/20200822/p2a/00m/0na/016000c#cxrecs_s

2020c. https://mainichi.jp/english/articles/20180719/p2a/00m/0na/004000c

NHK (2020): https://www3.nhk.or.jp/nhkworld/en/news/20200820_13/

Slate (2020): https://slate.com/technology/2020/07/climate-change-deaths-japan-2018-heat-wave.html

Worldometers (2020): https://www.worldometers.info/coronavirus/country/japan/


California Bushfires, August 2020

Paul Somerville, Chief Geoscientist, Risk Frontiers

Nearly 771,000 acres of largely unpopulated land have burned across California during the past week as dozens of lightning-sparked wildfires moved quickly through dry vegetation and threatened the edges of cities and towns. The fires have been most severe in the state’s northern and central regions, where about 600,000 acres have burned in the past week (Figure 1).

Evacuations surged on August 18 and 19 as authorities worried that high heat and gusty winds could cause the fires to spread rapidly. The resulting fires – and complexes of many small fires – have merged into major conflagrations in many parts of the state. By August 20, several of the major fires had more than doubled in size, in some cases jumping across major highways, as crews struggled to contain the blazes. By August 21, the two largest blazes, the SCU[1] and LNU[2] Lightning Complexes, had charred 340,000 and 325,000 acres respectively, becoming the second and third largest fires in California history (Table 1). The CZU[3] Lightning Fire forced the evacuation of more than 64,000 people, some of whom may not be able to return to their homes for weeks. Five people have died and about 1,000 structures have burned.

Figure 1. Left: Fire locations in California using Active Fire Data (hotspots) derived from the VIIRS for the last 7 days. Right: Satellite Image on August 19. Source: Washington Post.
Figure 2. Left: Fire in Napa, California. Right: Fire in Lassen County, California. Source: Washington Post.

The California wildfires, along with other blazes in the West, have sent a blanket of smoke across at st 10 states and southwestern Canada, with smoke extending over the Pacific Ocean as well (Figure 1, right panel). Air quality alerts are in effect for parts of California, where the tiny particles in the dense smoke are aggravating respiratory conditions and worsening preexisting health conditions that are already threatened by the coronavirus. The cloth masks that have now become a habit for many Californians when they venture outside are largely ineffective against the tiny smoke particles filling the air, and doctors recommend using N95 masks with vents. People are being asked to shelter in place, staying at home with their windows closed and ventilation systems set to recirculate air, which is difficult during a heatwave in areas such as San Francisco where many people do not have air conditioning.

A rare mix of ingredients came together in central and northern California to produce fast-moving, explosively growing wildfires that are powerful enough to create their own weather. Doppler radar revealed at least five tornado-strength rotational signatures inside the smoke plume in Lassen County, California. The record heat reached astonishing levels during the past two weeks as a massive “heat dome” parked itself over the West. On August 16, Death Valley, California, reached 130 degrees Fahrenheit (54 degrees Celsius). The combination of an intense, long-lasting heatwave, dry vegetation at the end of the summer, and a rare outbreak of August thunderstorms led to these blazes. Fueled by the heat, thunderstorms broke out on Sunday Aug 15 as a surge of tropical moisture pushed inland. The storms’ 20,000 lightning strikes (Fig. 3), including dry lightning storms, sparked more than two dozen blazes over a period of 3 days. An ancient stand of the world’s tallest trees has fallen victim to California’s raging wildfires. The CZU and SCU complex fires near Santa Cruz have ravaged Big Basin State Park, California’s oldest state park, some of whose giant redwoods are more than 50 feet around and 1,000 to 1,800 years old (Fig. 4).

Figure 3. Lightning storms in San Francisco and Healdsburg. Source: Washington Post.
Figure 4. Giant redwoods in Big Basin State Park. Source: Washington Post.

This is just the beginning of the state’s wildfire season, something that has been a constant threat during the past four years of blazes, some sparked by downed powerlines, that have set records for size and lethality. Despite the familiarity, the current fires and their speed and thick smoke have presented a new terror amid a global pandemic – poor air quality and concerns about evacuating masses of people to crowded shelters, and that some might not heed the warnings. Tens of thousands of people have been asked to evacuate and make difficult decisions about where to go. In the past, they might have stayed with friends or family, but now they need to calculate the risk of exposure to the novel coronavirus. Wherever people go, they are likely to face other hardships. California has been enduring a record-breaking heatwave that has prompted rolling blackouts because of high electricity demands for air conditioning and other uses. Most of the area is also experiencing severe or moderate drought.

In Santa Cruz and San Mateo Counties, south of San Francisco, about 48,000 people were ordered to evacuate because of a fire, part of the CZU Lightning Complex, that is threatening communities there. The blaze has already burned 50 structures. On the evening of August 20, the University of California at Santa Cruz was under mandatory evacuation and had declared a state of emergency.

The largest of the lightning-related fires was north of San Francisco, covering Napa and Sonoma counties. On August 20, that mass of fires, the LNU Lightning Complex, had grown to 219,000 acres and was uncontained. Approximately 30,000 structures were at risk of burning and 480 had been destroyed.

The blaze near Vacaville, known as the Hennessey Fire and part of the LNU Lightning Complex, has been one of the most destructive, burning down homes and claiming the life of a PG&E worker who was assisting first responders. This blaze burned down the La Borgata Winery and Distillery in Vacaville. Mandatory evacuations remained in effect for the north part of the city on August 20, and CalFire reported three additional civilian fatalities associated with the LNU Lightning Complex.

CalFire is at normal staffing levels, with approximately 12,000 firefighters working on August 21. Additional firefighters are being sought from other states and from Australia. In Central California, a pilot on a firefighting flight near Fresno died when his helicopter crashed.

Overall losses include 5 deaths, 64,000 people evacuated, over one thousand structures burned, 31,000 structures threatened, and approximately one million acres burned as of August 21.

The 2019/20 bushfires in eastern Australia were fought under dire conditions, but the presence of the coronavirus in California has made fire-fighting conditions there even more dire, especially those relating to evacuation.  There are 665,000 coronavirus cases in the state, growing by 5,000 a day, and 12,000 deaths, growing by 150 a day.

Table 1. 20 largest wildfires in California since 1932. Only 3 occurred before 2000. (Source: Updated from Cal Fire)


[1] SCU Lightning Complex Fire: Contra Costa, Alameda, Santa Clara, Stanislaus and San Joaquin counties

[2] LNU Lightning Complex Fire: Napa, Sonoma, Solano, Yolo and Lake counties

[3] CZU Lightning Complex Fire: San Mateo and Santa Cruz counties

Risk based earthquake pricing using catastrophe model output

Paul Somerville and Valentina Koschatzky,  Risk Frontiers

As the insurance market trends toward more analytical and data-driven decisions, insurers are continually exploring ways to rate risk better and more precisely. For the case of earthquake risk, this means an enhanced understanding of the relationship between event location, frequency, severity, how buildings respond to an event and the ensuing financial costs. The increased quantity, quality and granularity (resolution) of the available underwriting data and highly refined rating engines give insurers the opportunity to become extremely risk-specific in their pricing. Risk-based pricing – charging different rates depending on different risk characteristics of specific policies and in contrast to portfolio underwriting – leads to stability and confidence in pricing.  Risk based pricing aims to ensure that premium levels are commensurate with individual property risk profiles, with those in highly exposed areas experiencing a specific rate on the earthquake component of their coverage.  This seems to be a fairer and more equitable way of pricing risk. The ability to differentiate between perceived risk and actual risk affords insurers a better way to achieve their financial goals, allocate capital and meet client needs for coverage.

Several features of earthquake hazards and risks render them readily amenable to risk-based pricing.  First, the level of seismic hazard is not uniformly distributed across a country. New Zealand is an extreme example in which Wellington is located directly on a tectonic plate boundary having extremely high seismic hazard, whereas Auckland is remote from the plate boundary and has a seismic hazard level comparable to that of Australia (Figure 1).  However, even in Australia, the seismic hazard level also varies by an order on magnitude between relatively high levels in northwestern Western Australia, the Yilgarn region east of Perth, Adelaide, and southeastern Australia on the one hand and the extremely low levels in Queensland.

Figure 1. Peak acceleration maps for 1:500 AEP on Risk Frontiers’ Variable Resolution Grid for Australia and New Zealand.
Figure 1. Peak acceleration maps for 1:500 AEP on Risk Frontiers’ Variable Resolution Grid for Australia and New Zealand.

Second, the factors that increase the level of the hazard are well understood and mapped.  These include the presence of soils that amplify the level of ground shaking compared with that on rock, and the presence of saturated sands that can be liquefied during earthquake shaking, as occurred in Christchurch during the 2010-2011 Canterbury earthquake sequence.

Third, we are able to quantify the variations in building vulnerability to earthquake damage due to different building types, heights, ages of construction, and whether seismic building code provisions were used in design on a very specific basis.  G-NAF (Geocoded National Address File) is a geocoded address index listing all valid physical addresses in Australia. NEXIS (National Exposure Information System) is a database developed by Geoscience Australia containing building details for residential commercial and industrial buildings in Australia at a Statistical Area 1 (SA1) level.  There are 57,523 SA1 in Australia. These datasets allows wood, Mid-rise Steel, Concrete, and Reinforced Masonry and low-rise Unreinforced Masonry buildings damage ratios to be modelled and enable customised underwriting in Australia at the location, SA1 or postcode level. For New Zealand, the use of a variable resolution grid created using the Linz NZ street addresses database enables us to calculate the ground shaking hazard at a resolution as fine as 500 m while the liquefaction hazard is  calculated at the address level with a resolution of 16 m.

Finally, Risk Frontiers’ QuakeAUS and QuakeNZ models use a level of refinement in property damage estimation that is unique in the worldwide catastrophe loss modelling industry. Conventional earthquake loss estimation uses building fragility functions that are pre-computed using standard capacity curves for each building category of interest with a simplified representation of the building demand curve in response to ground-shaking. We instead account for the entire response spectral shape of the ground motion, which varies with many factors, including the earthquake magnitude, earthquake distance, and soil category at the risk location. Accordingly, our loss model dynamically calculates fragility curves for each building category at each site for each earthquake in the event set.  This produces building- and event-specific damages for each building category for each event, enhancing the accuracy and reliability of the loss calculation.

These four categories of information are combined to make detailed estimates of losses for each building that are then aggregated to obtain portfolio loss estimates.  However, it is an easy step to use this detailed information to quantify potential losses for any soil type or building category at any desired level of spatial resolution. For example, our model output can provide postcode level risk premiums (average annual loss AAL’s) for all of Australia and New Zealand for a nominal risk to estimate loss rate due to earthquakes for the following building modifiers, as shown in the example in Table 1:

  • Structure Type: Unknown, Wood/Light Frame low-rise, Steel Moment Frame mid-rise, Concrete Moment Frame mid-rise, Reinforced Masonry Bearing Walls mid-rise, unreinforced masonry low-rise. Mid-rise is defined as 4+floors, low rise as 1-2 floors
  • Year-built: pre-code (1980) and post-code
  • Damage calculated separately for buildings and contents
  • Separate estimates of direct damage and demand surge
Risk Premium
Postcode Structure Type Construction Date Building Contents
2294 Light Wood Unknown 74 64
2294 Mid-rise Steel Moment Frame after 1981 26 3
2294 Mid-rise Concrete Moment Frame before 1981 44 7
2294 Mid-rise reinforced Masonry Bearing Walls Unknown 35 25
2294 Low-rise Unreinforced masonry bearing Walls after 1981 226 58
2294 Unknown before 1981 111 71
2294 Low-rise Unreinforced masonry bearing Walls after 1981 141 58
2294 Low-rise Unreinforced masonry bearing Walls unknown 142 59
2286 Low-rise Unreinforced masonry bearing Walls before 1981 51 10
2295 Low-rise Unreinforced masonry bearing Walls before 1981 168 42
2291 Low-rise Unreinforced masonry bearing Walls before 1981 90 21

Table 1. Newcastle region risk premiums for building and contents with a nominal sum-insured. Earthquake risk based on location (postcode), construction type and year of construction can inform better underwriting decisions.

This bottom-up understanding of risk and pricing will also lead to better alignment of risk-premium and capital management.

Figure 2. By introducing earthquake risk pricing, insurers have an opportunity to align portfolio risk (left) and capital management with original premium rating and risk selection (right).
Figure 2. By introducing earthquake risk pricing, insurers have an opportunity to align portfolio risk (left) and capital management with original premium rating and risk selection (right).