Using Non-authoritative Sources During Emergencies in Urban Areas - Computational Approaches for Urban Environments - page 331

spatial statistical algorithms (Zhang et al. 2014 ). For example, points which identify
flooding (e.g., photos) are plotted, georeferenced, and then smoothed using a kernel
interpolation to create a GIS layer of estimated flood extent. This is a task performed
for each data source, resulting in multiple, individual flood extent layers. The
analysis is then performed on the fused layers by applying statistical or machine
learning algorithms to classify the data and identify anomalies.
Fusing data from multiple sources leads to an improved damage estimation and
an increased understanding of the sequence of events that leads to transportation
infrastructure failure. In this example, non-authoritative data are used in two
different scenarios:

1. Damage assessment during an event
2. Damage assessment after the event

It is assumed that ground truth data might not be available. The novelty of
this study is the development of a methodology that takes advantage of “citizens
as sensors” (Goodchild 2007 ) and of various other data, including remote sensing
and numerical models, not necessarily designed to be used during emergencies to
improve damage assessment. These non-authoritative or nontraditional sources are
used to create additional layers which augment traditional sources when they may
be lacking or incomplete. The result is shown in the bottom layer, where a flood
hazard map is generated. The resulting flood hazard map is then paired with a high-
resolution road network to create a road damage map.

14.2.1 Data Sources

14.2.1.1 Remote Sensing Data

High-resolution remote sensing data are routinely used to assess damage during and
after hazards, in both urban and rural areas. Two or more images are acquired for an
area showing the differences before and after the hazard.
The Civil Air Patrol, the civilian branch of the US Air Force, was tasked with
collecting aerial photos of the US East Coast following the impact of Hurricane
Sandy in October 2012. Within days of the storm making landfall, hundreds of
missions were flown by volunteers from Cape Cod, MA, to Cape May, NJ. From
these missions, thousands of aerial photos of the coastline were generated, including
those documenting heavily flooded areas.

14.2.1.2 Numerical Surge Model

Recent improvements in understanding the physics of storm surge combined
with rapid increases in High Performance Computing (HPC) power have led to