Log In
Access all customer product support, event response, and training in one place
LifeRisks PortalFind modeling tools based on best practice actuarial techniques and medical science
Miu PortalExplore analytics and risk insights for the alternative capital market
Insurance Solutions
Formerly Moody’s RMS
Christopher Allen is senior analyst – model development, working within the Event Response team at RMS, and is based in London. He joined fellow employees from RMS and RMS clients on our annual Impact Trek in Nepal during March this year. This is Chris’s account of his time in Nepal.
On Wednesday, March 21, eight somewhat-still-jetlagged RMS Impact Trekkers left the Summit Hotel in Kathmandu bound for the village of Dhunkharka, three hours’ drive southeast of the capital. We were going to see some of the retrofitting work that was being done by Build Change, a social enterprise partner of RMS that aims to build local capacity for safer construction practices. As we weaved our way through Kathmandu traffic (a chaotic affair at the best of times) we noticed several of the characteristics of the capital’s buildings that had been pointed out to us by Build Change over the past few days: soft story, three-floor brick masonry, new construction sprouting up with reinforced concrete columns, the occasional ground floor still occupied by goats or buffalo…
After about an hour we had escaped the worst smog of the capital and were heading into the hills, passing brick kiln after brick kiln. Not long after, we turned off the paved road and our Jeeps began their bumpy ride on a dirt track that wound its way higher and higher up the dusty hillside. At around 10.30 a.m. we stopped in a small village for lunch. This was, naturally, several servings of daal bhat, a meal of rice and lentil soup that most Nepalis eat twice a day as a staple meal. While we were waiting for the food to arrive, I had the opportunity to practice some very hastily learnt Nepali with some local schoolkids: “Kasto cha?” (How are you?) and “Timro naam ke ho?” (What is your name?)
After eating we walked a short distance down the hill to inspect a model home that Build Change had constructed. Different structural elements had been color coded (see picture below) to make it easy for the design to be copied: the sill band, the through stones, concrete spacers etc. Nearby was a billboard which explained how each feature improved the building’s performance in an earthquake. The building was not left empty though, it was being used as an antenatal center where women could go several days before giving birth, being just opposite the medical clinic.
We returned to the Jeeps and continued up a steep rocky lane. We didn’t get far though, after about five minutes the track turned into a bog and we had no choice but to abandon the vehicles and walk the remainder of the way to Dhunkharka. This turned out to be a far more pleasant experience than driving however, as we could take in the views, the sunshine, and chat to the Build Change staff with us, Kiran and Dinesh. After about a twenty-minute climb we arrived in the village, which was little more than a number of scattered houses and homestays.
We could see earthquake damage to some of the buildings, and one had even been demolished as it had been deemed unsafe. Kieran took us further uphill to visit the house of an elderly couple that Build Change had retrofitted to make safe after the earthquake. They were kind enough to let us inside, and we inspected the “soft retrofit” wooden columns that had been installed all the way from floor to rooftop to provide several spines to the three-story home. The elderly couple were happy to talk to us about the earthquake and how frightening it had been, and animatedly told us their plans to turn their house into a homestay in order to raise some more money.
Our next stop was the Build Change local training center. This was a home in a neighboring village where local builders and laborers could learn how to retrofit, step-by-step. Build Change has just begun teaching a structured competency-based course, where attendees are taught together then assessed individually on each element of the course, for example, correctly assembling the joists or replacing the gable wall. For me, the most fascinating aspect of the retrofitting process was the use of huge ratchet bands (see picture below) around the entire outside of the building in order to hold it together while the retrofit was being completed.
It was clear, however, that Build Change did not just concern themselves with engineering but thought much more widely about how to understand and engage with the whole community where they worked. As an example, laborers who had to miss a day’s work in order to attend the competency-based training were paid a per diem.
All too soon it was time to leave and we piled into our Jeeps again, somehow managing to doze off on the ride back despite the rocky track out of Dhunkharka and remaining sound asleep as the cars entered Kathmandu, swerving to avoid yet another cow ambling across the ring road.
Late May 2019 was a startlingly active period for severe convective storms (SCS) in the U.S., even after considering that May is typically one of the most active months of the year. Until about halfway through the month, the number of tornadoes being reported was around average, but after a major outbreak starting in mid-May this number shot up, bringing the year-to-date total to 1,017 tornado reports. This count is only surpassed by the extremely active years of 2008 and 2011 (Figure 1). This year’s late May outbreak was also unusually long: by the end of Wednesday, May 29, at least eight tornadoes had been experienced each day across a record-breaking 13 consecutive days, according to preliminary data from the National Weather Service (NWS). The previous record was set in 1980, after 11 consecutive days with at least eight tornadoes. Figure 1: Cumulative counts of tornado reports. 2019 (up to 30 May) shown in black. 2016-18 data are not plotted. Source: https://www.spc.noaa.gov/wcm/The reason this latest outbreak lasted so long was due to a particularly persistent set of atmospheric conditions favourable for SCS. A combination of a stubborn high-pressure system over the Southeast U.S. and a stationary and unusually cold trough over the Rocky Mountains, saw warm, moist air drawn from the Gulf of Mexico into the central U.S., and a strong jet stream aided the formation of supercell thunderstorms, which can generate very powerful tornadoes. Hailstone measuring 5.5 inches found at Wellington, Texas, on May 20, 2019. Image credit: Twitter/@NWSAmarilloThe outbreak spawned tornadoes across a wide area of the U.S. (Figure 2), but most have occurred in the southern Great Plains (so-called “Tornado Alley”) and the Midwest. It has not only been tornadoes causing damage, the severe weather also generated large hail – with a 5.5 inch (14 centimeter) diameter hailstone (see above) reported in Texas, and strong straight-line winds. Figure 2: Tornado reports by U.S. County (20-30 May 2019). Data source: NOAA (https://www.spc.noaa.gov/climo/reports/)SCS is a major contributor to annual insured losses in the United States. RMS modeling shows that average annual loss (AAL) from SCS roughly equals that from hurricanes, at around US$17 billion. Unlike losses from hurricanes, SCS losses (especially hail) accumulate over the course of a year; however, major outbreaks can make the difference between an “average” year and a significant impact to the insurance industry. Prolonged and damaging outbreaks also increase the likelihood that losses will flow through to reinsurance layers. The most significant past tornado outbreaks, for example, the tornadoes that hit Tuscaloosa, Alabama and Joplin, Missouri in 2011, have cost single-digit billions of insured loss (USD). One particularly interesting characteristic of the late-May 2019 SCS outbreak, however, is that the damage could have been much worse. Meteorological records have been broken but – for the most part – the tornadoes avoided major cities. That said, it will take some time before all of the claims (hail, tornado and wind) have been verified and processed, so it is too early to know what the cost of the outbreak to the insurance industry will be. It is worth noting that the SCS outbreak was accompanied by significant flooding in Arkansas, Oklahoma and Missouri, but since flood damage is not typically covered by private insurance in the U.S., if the properties were not covered by the National Flood Insurance Program (NFIP), these costs will be borne by individual home and business owners. Below are shown some statistics and charts covering this remarkable outbreak. Unless otherwise stated, they cover the period 20-30 May 2019: Number of Confirmed Tornadoes in the U.S. (May 16–30): ≥ 228 22 of these have not yet been given a strength rating by the National Weather Service (NWS). More tornadoes may be confirmed by the NWS as their surveys continue. Number of EF5 Tornadoes: 0 There have been no tornadoes of the highest category on the Enhanced Fujita (EF) Scale. Number of EF4 Tornadoes: 3 EF-4 in Dayton, Ohio EF-4 in Linwood, Kansas EF-4 hit Marshall, Oklahoma …
Chris is the product manager for the RMS North America Severe Convective Storm (SCS) and Winter Storm models. He has a doctorate in Climatology from Oxford University, where he was a College Lecturer.
Apart from SCS and winter storm, he is particularly interested in hurricanes and still gets excited by his old PhD topic, Saharan dust storms. Chris is a Certified Catastrophe Risk Analyst.
