Spatial and temporal distribution of hazards Flashcards
Natural hazards are
Natural Hazards are naturally-occurring physical phenomena caused by either rapid or slow onset events having atmospheric, geologic and hydrologic origins at a global, regional, national or local scale. They include earthquakes, volcanicc eruptions, landslides, tsunamis etc (UNESCO, 2013)
Natural disasters are
“Natural disasters are the consequences or effects of natural hazards, but natural phenomena do not automatically spell disaster” (UNESCO, 2013)
Importance of understanding physical characteristics
Identifying common features – generalizations
Mechanistic controls – potential for damage/destruction
Linking physical processes to mitigation measures
Categorization of hazards - typologies
Typologies provide a useful framework for identifying similarities and making generalisations about hazardous events
Typologies also promote sound management practices.
Origin based classifications
Atmosphere/hydrosphere, lithosphere, and biosphere (Chapman, 1999)
Endogenous (forces from within the earth), exogenous (focus above the earth surface)and anthropogenous
EM-DAT – natural and technological (Hydro-meteorological and geophysical)
Primary and secondary hazards
primary e.g. earthquakes, storm surges, volcanic eruptions etc. Secondary hazards e.g. landslides, tsunamis, pyroclastic flows etc.
Magnitude
Measure of strength/force
Comparison of extreme events in space and time
Applicable to all hazards – baselines
Magnitude Limitations
Bad indicator of impact and hazardousness
Scale(s) of measurement –
what do they mean?
Houghton et al (2013) measured
During explosive eruptions of Kilaueau in 2008 we constructed the first time deposits of bulk volumes to demonstrate exponential thinning from the vent
Houghton et al (2013) VEI
The VEI is increasingly being used as a measure of magnitude of explosive eruptions
Houghton et al (2013) The 2008 eruption deposits demonstrate
A problem for the use of VEI, as originally defined, which classifies small, yet ballistic producing explosive eruptions at Kilaueau as non explosive
Musson et al (2010) scales to measure earthquakes
Numerous macroseismic scales as an index of shaking with the number of important scales adopted much smaller, maybe 8.
Musson et al (2010) Importance of scale
The extent to which a scale guides the user to arrive at a correct assessment of the intensity is a measure of the quality of the scale
Musson et al (2010) a useful scale must consider
Applicability - diagnostics; consistency - are diagnostics equivalent; discrimination - diagnostics might not be an expression of intensity; number of degrees - depends on ability to resolve intensities; regularity - poor practice and isoseismal maps; reliability - seek a method that reduces subjectivity.
Doswell et al (2009) The F scale
Was originally formulated as a peak wind speed scale for tornadoes; it has been implemented using damage to estimate wind speed
Doswell et al (2009) The EF scale
Recently, the F scale has been replaced in the US by an official system for rating tornado intensity
Doswell et al (2009) Conclusion
The adoption of the EF scale may have been premature, especially if it is to serve as a model for how to rate tornadoes outside the US
Duration
Hazard type – effects on point/area (space)
The relationship between duration and hazard planning/management
Duration and period of onset
Duration and scale of impact (area affected)
Duration: Boxing Day Tsunami
Cause – megathrust earthquake – Indian plate subducted by Burma plate
Magnitude 9.0 – revised estimate – 2005
Hypocentre, N Sumatra, 1300km rupture
Teletsunami – vertical seabed displacement
Travel characteristics
Travel time 15min-7 hours
Temporal distribution - frequency
Qualitative or quantitative description The relationship between recurrence and magnitude The role of historical records Standardisation of scales Technological advances
Smith et al (2010) hurricane activity increase
North Atlantic hurricane activity has increased substantially since the 1970s, but whether this is attributable to natural internal variability or external forcing has not been resolved.
Smith et al (2010) hurricane frequency predictable?
Hurricane frequency is potentially predictable, because climate models can directly simulate year-to-year variations in Atlantic tropical storm frequency, if forced by observed SST’s.
Smith et al (2010) Found
physically based model evidence of externally forced changes in hurricane frequency, albeit from a single modelling system. Smith’s results show that predictions of hurricane frequency are viable beyond the seasonal scale, and further elucidate causes of hurricane variability.- decadal predictions using GCMs
Frequency: Tropical cyclones: IPCC Special Report (2012)
Detection of trends (frequency, intensity and duration) remains a challenge
Past records – heterogeneous due to observing technology and reporting protocols
Regional trends in North Atlantic – fidelity disputed
Steady global TC frequency – inter-annual and multi-decadal trends within basins
Projected increase in most intense storms
Relative simplicity of seasonality
The realm of hydro-meteorological hazards
Seasonality/periodicity
Predictability – timing and location
The relationship between seasonality and spatial location
Management hampered by uncertainties surrounding frequency and intensity
Limitations – e.g. UK snow and EU floods
Moving Seasons - Westerling et al (2006)- wildfire increase
Western US wildfire activity is widely thought to have increased in recent decades
Moving Seasons - Westerling et al (2006) - Database
Compiled a comprehensive database of large wildfires in western US forests since 1970 and compared it with hydroclimatic and land surface data
Moving Seasons - Westerling et al (2006) - Findings
Wildfire has increased in the mid 1980s - the greatest increase occured in mid elevation Northern Rockies where land use has relatively little effect on fire risks
Importance of diurnal factors
Thunderstorms and tornadoes – physical controls in specific areas
Time of day – important , regardless of hazard type
Space: the barrier to generalisation
Multiple physical parameters of a single hazard provide a stronger basis for comparison
Characterisation of hazard through characteristics = problematic – variation in properties at different spatial scales
Areal extent
- the area affected – hazard type, environmental constraints
Spatial distribution
– the space in which hazards occur
Berz et al (2001) Munich Re’s Geoscience research group
Published its first world map of natural hazards in 1978
Berz et al (2001) Revised map
In 1998 the map was revised with all first time recorded data recorded and analysed using GIS, background info of earthquakes, volcanism, windstorm, floods etc added in and hazard info has been stated as numbers so it can be checked
Peduzzi et al (2005) - there is a call for
An improved methodology allowing comparison of natural hazard impacts at a global level
Peduzzi et al (2005) - data needed
In order to associate reported impacts with affected elements and socio economic or geophysical contextual parameters, geographical location and extent of hazards is needed.
Peduzzi et al (2005) - results
Presents an improved automated procedure for mapping a larger disastrous hazard events using GIS. Up to 82% of the events and 88% of the reported victims were georeferenced.
Dixon et al (2011) - study suffers
This study suffers from shortcomings associated with the tornado database. By avoiding separation by F scale and by using local tornado days rather than raw tornado events, major potential problems with the dataset have been avoided
Dizon et al (2011) - KDE method
The KDE method provides a robust assessment of tornado threat in a particular area, but the calculations are limited by the quality of the recorded tornado paths. – it is assumed that an employed kernel radius (40.25km) is sufficient to mask any spatial areas, but some events that travelled significant distanced might have been reported as single-point touchdown
Dixon et al (2011) - tornado-day density and dixie valley
Statistical analysis shows that regions with similar tornado-day densities are located throughout the Great Plains, the Corn Belt, and the Deep South without areas of statistically significant differences separating them.
It is possible that this line of relatively little tornado activity is partially responsible for the emergence of “Dixie Alley” as a separate region.
Brooks et al (2013)
showed that spatial patterns vary dramatically with seasons
Dilley et al (2005) - world bank study
Revealed that more than half the world’s population were exposed to one or more natural hazards
Bower (2011) trend in hydro met data
22 studies of economic losses data over at least 30 years for hydro met disasters - in 14 cases the normalized data showed no trend.
Garner and Huff (1997) reporting
Medium reporting shows an excessive concentration on the emergency phase of a disaster, especially if striking images of distressed victims are available
NOAA - reporting time of tornadoes
the average amount of time between a tornado warning and the arrival of a storm is about 13 minutes
National Geographic article, March 2015
We don’t understand how tornadoes die. Brooks says tornadoes tend to follow the general movement of the thunderstorm they are associated with, but the route can be erratic.
