NG33A-0882 1340h
Forest Fire Burn Areas Modeled as Self-Similar Criticality
Forest fire burn areas in parts of western Canada have cumulative frequency-area distributions that are well-described by a power law or an upper-truncated power law. This power law scaling extends over as many as five orders of magnitude and is observed for different geographical regions and for different time intervals. The observed scaling exponent varies both geographically within and between provinces and temporally between annual records. The temporal variability decreases at the decadal scale, suggesting that decadal distributions may be useful for long term forecasting within a geographical region. The traditional Self-Organized Criticality (SOC) forest fire model does not produce the range of scaling exponents observed for natural fires. The model of Self-Similar Criticality (SSC), which has been used successfully to model hotspot seamount formation, replicates the cumulative frequency-area distributions observed for forest fires in western Canada.
NG33A-0883 1340h
Critical behaviour of the seismic precursors of a cliff collapse
We analyse the statistical pattern of seismicity before a 1-2 103 m3 chalk cliff collapse on the Normandie ocean shore, Western France. About 500 seismic events, in both 40 Hz-1.5 kHz and 2 Hz-10kHz frequency range, have been recorded during the six months before the collapse, and more than 200 during the two hours preceding the collapse. The collapse occurred within a seismic network of five microseismic stations, in such a manner that the central station was located at few meters of the failure surface. This allows us to investigate the properties of the seismicity emitted by the damage process leading to the cliff collapse. The study focused on the statistical properties of the seismic events. The event size distribution displays a power law distribution ($N(E)\propto E^{-b}$) on more than 3 magnitude orders with an exponent of 0.55. Such a distribution can be compared to the Gutenberg-Richter law observed for crustal earthquakes. During the last hour before the collapse, we observed a continuous decrease of b-value until the collapse occurred, indicating that the proportion of large events increases toward the time to failure. We show that, contemporary to this exponent decrease, a power law acceleration of seismicity rate and energy is defined on 3 order of magnitude, within 2 hours from the collapse time. We discuss these results, in the context of brittle failure and tertiary creep models. Our analysis of this first seismic monitoring data of a cliff collapse suggests that the thermodynamic phase transition models for brittle rupture may apply for this cliff collapse. We suggest that the case study presented here is one of the first (even the first) case showing a power law acceleration before a cliff collapse. It open new routes both to monitor and to understand the physics rock slope and landslide instabilities.
NG33A-0884 1340h
Yet More Lessons From Complexity. Unity the key for Peace.
The last few decades have witnessed the development of a host of ideas aimed at understanding and predicting nature's ever present complexity. It is shown that such a work provides, through its detailed study of order and disorder, a suitable framework for visualizing the dynamics and consequences of mankind's ever present divisive traits. Specifically, this work explains how recent universal results pertaining to power-laws, self-organized criticality and space-filling transformations provide additional and pertinent reminders that point us to unity as an essential element for us to achieve peace.
NG33A-0885 1340h
Landslides in Vibrating Sand-Box; Preliminary Results Reporting Types of Slope-Failure and Apparent Frequency Magnitude (Area) Power Law Relations.
It is recognized that hazardous natural phenomena such as earthquakes, forest fires and landslides often follow a power-law frequency-magnitude relations. Naturally occurring landslides populations, both seismic and hydrologically triggered, show non-cumulative power law frequency-magnitude (area) relations with slope of 2.3-3 for the large landslides part of the population. Numerical simulations of sand pile avalanches obtain a non-cumulative frequency-size distribution which also follows a power-law but with a slope of 1. In this work we study the nature and area distribution of slope failure triggered in a vibrating wet (1%wt) sand box. We used a 28 cubic cm box with sand pile crest resting on the top of one inner face and foot on the opposite face base. Initial slope angle was about 50deg, vibrating frequency 10Hz and individual test duration lasted a few minutes. Three different accelerations directions were tested: vertical, slope perpendicular, and normal horizontal accelerations. Acceleration magnitudes ranged from 0.1 to 1.2g. Slope performance was continuously recorded using a digital video camera. We observed that vertical vibrations larger than 1.0 g, induced mainly a few centimeter wide block-slides and toppling from a step like scarp that migrated up the slope. Block sliding rate was approximately one every few seconds. Final slope cross-section is S shaped with normal faulting at its crest. Final slope angle was about 35deg. Lower accelerations or lower initial slope angles yielded only surface grain flow. Horizontal shaking yielded different behavior: Above a threshold acceleration (0.5g and 0.7g for shaking parallel and normal to slope direction, respectively), surface flow occurred initially. It was followed by a box-wide slump, which first remained coherent and then progressively disintegrated. Lower accelerations or initial slope angle yield only surface grain flow. Finally, the upper surface areas of tens of block-slides induced in the above described vertical vibration tests were analyzed. The blocks cumulative area distribution shows power law relation with slope of about 1 for blocks with area smaller than 1sq-cm. We find a previously unknown control on style of slope failure: the acceleration direction and amplitude.
NG33A-0886 1340h
Longshore sediment transport from Maine to Florida: Comparison of literature compilation to model results based on WIS hindcast deep-water data
We examine the regional longshore sediment transport pattern of the seaward coast of the United States and Gulf of Mexico from northern Maine to Tampa Bay, Florida. A regional interpretation of longshore transport compiled from published field and modeling studies is compared to gross and net potential sediment transport directions and rates determined with a model using wave information study (WIS) hindcast data. The approach will assess the validity of using WIS hindcast data to determine longshore transport rates and directions. Using an approach herein called the WIS deep-water method, the WIS deep-water wave characteristics are used in the equations of Ashton et al (2003) to model the direction and rate of longshore sediment transport at roughly 25 km intervals along the coast. The derived transport directions, including the indicated location of nodal zones, generally agree with published field studies, although there are some local inconsistencies, particularly near inlets, shoals and irregular bathymetry. Trends in longshore transport rates, such as increases and decreases in gross transport rates, are well represented by the potential transport rates derived by the WIS deep-water method. The discrepancies between the published field studies and WIS results are apparently primarily due to assumptions in the WIS model, such as assuming shore-parallel bathymetric contours. We conclude that where carefully conducted long-term field studies are available they are the preferred means of determining accurate longshore sediment transport rates and directions. When other data are not available, the WIS deep-water method is found to provide an accurate regional determination of sediment transport directions and relative magnitudes for roughly 70 percent of the study area at a fraction of the effort and cost.