ArticlePDF Available

Post-disaster damage assessments as catalysts for recovery: A look at assessments conducted in the wake of the 2015 earthquake in Nepal


Abstract and Figures

In the wake of large earthquake disasters, governments, international agencies, and large nongovernmental organizations scramble to conduct impact and damage assessments that help them understand the nature and scale of the emergency in order to orchestrate a complex series of emergency, response, and recovery activities. Using the Gorkha earthquake as a case study, this research seeks to provide greater clarity into the types of post-disaster damage assessments, their purposes, and their potential as catalysts for critical recovery activities. We argue that damage assessment methodologies need to be tailored to the diverse information needs in post-disaster contexts, which vary by user group and change over time. This research builds upon the authors' direct experience supporting the government of Nepal in the Post-Disaster Needs Assessment (PDNA) process, support with the rapid visual inspections conducted by the National Engineering Association, and interviews with humanitarian organizations who conducted damage assessment in Nepal.
Content may be subject to copyright.
The Professional Journal of the Earthquake Engineering Research Institute
This preprint is a PDF of a manuscript that has been accepted for
publication in Earthquake Spectra. It is the final version that was
uploaded and approved by the author(s). While the paper has been
through the usual rigorous peer review process for the Journal, it has
not been copyedited, nor have the figures and tables been modified for
final publication. Please also note that the paper may refer to online
Appendices that are not yet available.
We have posted this preliminary version of the manuscript online in
the interest of making the scientific findings available for distribution
and citation as quickly as possible following acceptance. However,
readers should be aware that the final, published version will look
different from this version and may also have some differences in
The DOI for this manuscript and the correct format for citing the paper are
given at the top of the online (html) abstract.
Once the final, published version of this paper is posted online, it will
replace the preliminary version at the specified DOI.
Post-disaster damage assessments as catalysts
for recovery: A look at assessments conducted
in the wake of the 2015 earthquake in Nepal
David Lallemant,a) M.EERI, Robert Soden,b) Steven Rubinyi,c) Sabine Loosd),
Karen Barns d), Gitanjali Bhattacherjee d)
In the wake of large earthquake disasters, governments, international agencies and
large NGOs scramble to conduct impact and damage assessments that help them
understand the nature and scale of the emergency in order to orchestrate a complex
series of emergency, response and recovery activities. Using the Gorkha earthquake as
a case study, this research seeks to provide greater clarity into the types of post-disaster
damage assessments, their purposes, and their potential as catalysts for critical recovery
activities. We argue that damage assessment methodologies need to be tailored to the
diverse information needs in post-disaster contexts, which vary by user group and
change over time. This research builds upon the authors’ direct experience supporting
the government of Nepal in the Post-Disaster Needs Assessment (PDNA) process,
support with the rapid visual inspections conducted by the National Engineering
Association, and interviews with humanitarian organizations who conducted damage
assessment in Nepal.
The magnitude 7.8 earthquake of April 25th 2015 and large aftershock on May 12th (7.3
magnitude) were the largest to affect Nepal since the 1934 great Bihar earthquake (estimated
8.0 magnitude). The 2015 events resulted in nearly 9,000 fatalities and 22,000 injuries. An
estimated 8 million people (a third of the Nepalese population) were directly affected, and
over half a million homes destroyed or damaged (Government of Nepal 2015). While already
underway, the recovery from a national disaster of this scale will take many years. This paper
a) Earth Observatory of Singapore, Nanyang Technological University, Singapore | Asian School of the
Environment, Nanyang Technological University, Singapore
b) University of Colorado, Boulder
c) Oxford University, Oxford, UK
d) Stanford University, Stanford CA
looks specifically at efforts to assess damages to buildings in the housing sector resulting
from the earthquake in Nepal. We document the multiple housing damage assessment
programs conducted in Nepal, their purposes, and their impact as catalysts for recovery.
While several assessments are described in broad terms, the paper additionally provides an
in-depth study of the Earthquake Household Damage and Characteristics Survey (EHDC)
conducted as part of the national Earthquake Housing Reconstruction Project (EHRP). The
paper concludes with a set of recommendations based on lessons learned from this damage
survey program in Nepal. As such, the paper aims to inform the design of damage
assessments in future disasters.
Five types of damage assessments are discussed in this paper: (1) modeled earthquake
damage estimates, (2) the post-disaster needs assessment (PDNA), (3) remote-sensing based
damage assessments, (4) rapid visual engineering building safety assessments, and (5)
recovery oriented housing damage surveys. This list is not exhaustive but represents the
range of housing damage assessment programs conducted after large earthquakes, for
different purposes and with different methods. Since damage assessment practices in each
disaster build upon a body of knowledge and experiences from previous disasters, we first
describe these damage assessments in general terms before discussing their specific
implementation in Nepal.
Within hours following a potentially damaging earthquake, the United States Geological
Survey (USGS) triggers the Prompt Assessment of Global Earthquakes for Response
(PAGER). PAGER is an automated system to estimate the impact of earthquakes. It uses the
earthquake source characterization to estimate shaking intensity across a region based on
ground-motion prediction equations and interpolation of the estimated intensity with
recorded intensity from seismic stations and people’s reporting through USGS’ “did-you-
feel-it?” online survey system. The PAGER system relates each level of shaking intensity to
estimated fatality and economic losses, calibrated to past earthquake events in the region in
question (Jaiswal & Wald 2013; Earle et al. 2009). This model is used to produce earthquake
impact summary reports, describing the probabilistic distribution of fatalities and financial
losses, population exposed to various levels of shaking, and shaking intensity maps.
The PAGER impact estimate is a good example of what we will call “modeled
earthquake damage estimates”. Since these estimates are released before eyewitness reports
can be put together, they are not based on observed impact. Hence a distinction is made here
between damage assessment and damage estimates. A damage assessment is based on direct
observation of damage, while a damage estimate is based on modeled expectations of
damage. Besides USGS-PAGER, several other prominent models have been developed:
QUAKELOSS produced by the World Agency of Planetary Monitoring and Earthquake Risk
Reduction; the Global Disaster Alert and Coordination System (GDACS) produced by the
Joint Research Center (European Commission), and others. Typically, these models combine
an estimate of the intensity of ground shaking (hazard) with an estimate of the population
exposed to that shaking (exposure) and their susceptibility to the shaking (vulnerability).
Model accuracy varies depending on assumptions and the context of the event. The main
sources of uncertainty in the impact estimate are: density of the seismic network (which
constrains the uncertainty in the shaking intensity maps); accuracy of exposure
characterization (distribution of population and building types); availability of locally
specific vulnerability curves (that describe local building construction); and time/occupancy
model (that describe temporal fluctuation in building occupation). The accuracy of models
therefore increases in contexts that are data rich and have had recent seismic activity.
Modeled earthquake damage estimates are the first broad impact summary produced
following earthquake events. In their absence, early indications of the scale and distribution
of earthquake impact are based solely upon the magnitude of the event and eyewitness
testimony. Modeled earthquake damage estimates typically serve as the triggering
mechanism for numerous response activities. For instance, the Office of U.S. Foreign
Disaster Assistance (OFDA) and its Disaster Assistance Response Team (DART) is typically
mobilized in response to the USGS PAGER impact assessment (Earle et al. 2009). Similarly,
humanitarian organizations, government institutions, development organizations, and the
media rely on rapid damage estimates for mobilization, pre-positioning, stockpiling and
operations planning even before field-based data is made available.
Modeled earthquake damage estimates were particularly important following the 2015
Gorkha earthquake, because the earthquake affected a large area (one third of Nepal), and in
particular remote mountain communities. USGS released the PAGER report the same day as
the earthquake and updated it regularly as the earthquake rupture and shaking intensity model
was updated. Nepal’s National Society of Earthquake Technologies (NSET), a Nepali
institute that provided earthquake impact and engineering information, relied on the USGS
“Shakemap and PAGER for their impact summary. Printed posters of the PAGER summary
could be seen on the walls of their main office in Kathmandu and linked from their website.
A group from the Stanford Urban Resilience Initiative (including the authors) produced an
impact estimate that was shared and presented to the World Bank headquarters and country
office team in the week following the earthquake (Binns 2016). This impact estimate was
used to scope assistance needs and as a communication tool with other agencies.
In 2008, the World Bank, European Commission and United Nations Development
Group signed the Joint Declaration on Post-Crisis Assessments and Recovery Planning
(GFDRR et al. 2013). The Declaration is an agreement to collaborate on a common approach
to assessing disaster impact and mobilizing financial assistance for disaster recovery through
the Post-Disaster Damage and Needs Assessment (PDNA). Launched by request and led by
the affected country government, the PDNA brings together national and international
stakeholders to collaboratively assess the disaster impact, the needs for recovery and the
strategy for its implementation. The output of the PDNA process is a consolidated multi-
sector assessment report to guide short-term and long-term recovery planning. Importantly,
the damage and needs figures presented in the report provide the basis for requests for
international development assistance. One of the core elements of the PDNA is the
assessment of damage to infrastructure and physical assets, organized by sector (e.g. housing,
health, cultural heritage, transportation, etc.) (GFDRR et al. 2013). The data collection and
analysis consists of a desk review based on secondary data and baseline pre-event data,
followed by interviews with key stakeholders and field assessment samples. The guidelines
to the PDNA explain that:
In many countries it would be difficult to collect data on the impact of disasters owing to
a lack of systematic data collection on the part of the relevant agencies or lack of access to
affected areas due to logistical, administrative or security constraints. In such a situation, the
primary data could be collected through limited, specifically designed surveys applied to
statistically relevant sample groups, depending on the availability of time and resources. The
results of such small sample surveys would be extrapolated to derive an estimate of the total
picture of the damages, losses, and needs. […] It is always preferable to have an extrapolated
estimate that can be used for assessment rather than having no data or an elaborate database
which is excessively time and resource-intensive. […] A number of information-collection
methods are used to assess the impact of disasters […]: Focus group discussions; Interviews
with livelihood groups; Interviews with key informants; Household visits and interviews;
Participant observation in the field; Household surveys; Maps and satellite imagery. (GFDRR
et al. 2013)
Hence the PDNA is often an ad-hoc process of identifying available data, assessing its
reliability, extrapolating it where gaps exist, and conducting limited validation through
sampled field sites. While the PDNA is an influential document, no systematic study has
been conducted to assess accuracy of the findings based on more in-depth damage surveys.
The PDNA for the Gorkha earthquake was launched on May 15th 2015 (3 weeks after the
main earthquake) and lasted till June 24th 2015. It was led by the Nepalese Government
National Planning Commission (NPC), supported by a core PDNA secretariat consisting of
representatives from the Asian Development Bank, the European Union, the United Nations,
the World Bank, and the Japan International Cooperation Agency. Over 250 government
officials and external experts from 30 development partner agencies were organized into 23
thematic groups within four sectors: Social, Productive Sectors, Infrastructure, and Cross-
cutting. We focus here on the Housing Sector assessment. The information presented is based
on the first authors experience as the co-lead for the housing sector PDNA working group.
The working group for the housing sector PDNA was made up of nearly 30 people,
consisting of Nepalese government officials, representatives from multilateral and bilateral
donor agencies and civil-society organizations. The goal of the working group was to
produce a table summarizing the impact of the earthquake on the housing sector in Nepal and
the needs for resilient housing recovery. It was organized as shown in Table 1.
Table 1: Nepal PDNA Housing Sector Assessment summary organization.
Houses Fully Destroyed (by housing type)
Houses Partially Destroyed (by housing type)
Household Goods Destroyed
Demolition and Rubble Removal
Loss of Rental Income
Housing reconstruction (to higher seismic standards)
Housing repair and retrofit
Transitional shelters
Others (settlement planning, training, etc.)
The number of buildings fully and partially destroyed per district are the primary information
needs in this sector, as they account for the vast majority of total damages. Most other impact
measures (i.e. losses, transitional shelter needs, and demolition and rubble clearance needs),
are then extrapolated from these primary figures.
It would not have been possible to complete a comprehensive housing damage survey
within the timeline on which the PDNA needed to be delivered, roughly one month after the
earthquake. However, the Ministry of Home Affairs (MoHA) mobilized local police officials
to gather available information on fatalities and building damage in each affected district.
These figures were provided online through the Nepalese Disaster Risk Reduction
Portal ( Despite questions about the reliability of these figures
(discussed later), they became the main information source for the housing component of the
PDNA. Since the replacement cost of buildings varies significantly by building type, it was
necessary to disaggregate the MoHA damage figures by building type. Nepal’s detailed
building census, conducted in 2011, was used to define three main housing typologies (low-
strength masonry, cement-mortared masonry, and reinforced concrete) and their distribution
at the district level. Rather than extrapolating building damage figures across building types
according to their census distributions, building fragility curves developed by NSET were
used to back-calculate the expected distribution of damage for each building type. Field visits
were conducted to estimate the average replacement and repair-cost for each building type.
While it is clear that the resulting damage figures have significant uncertainty due to the
many assumptions embedded in the analysis, this is quite typical of PDNAs. The time
between a disaster event and the launch of the PDNA is typically too short for extensive
damage surveys to be conducted. Moreover, PDNAs are conducted most often in contexts of
limited institutional capacity for systematic data-gathering. Hence PDNA participants are
encouraged to collect any data available, and use judgment to extrapolate them to meet the
needs of the PDNA. In the case of Nepal, there was significant discussion surrounding the
reliability of the damage figures provided by the MoHA survey. In particular, questions arose
when damage figures for a specific district matched the census figures for total number of
households exactly, suggesting that 100% of the 66,636 buildings in that district were
destroyed (96%) or partially destroyed (4%). Similar patterns were present in other districts.
Despite these issues, it was ultimately decided that these figures were the most reliable
available and, because they originated from MoHA, had the advantage of authoritativeness.
The use of satellite images, radar and other remote-sensing technologies has become
standard practice for rapid post-disaster situational awareness. Remote-sensing technologies
are used extensively and for a variety of post-disaster and humanitarian purposes: to produce
base-maps for unmapped areas affected by disaster, to identify fault traces, to identify
landslides or other large land perturbations, and increasingly to assess damages (Voigt et al.
2007; Joyce et al. 2009; Gnyawali & Adhikari 2016; Dong & Shan 2013). Following
modeled damage estimates, remote-sensing based damage assessments typically provide the
next fastest estimates of damage following disaster. Instead of relying on modeled damage,
they use either manual or automated approaches to measure impact through analysis of
satellite images or other Earth Observation (EO) technologies such as radar.
Automated techniques most often rely on change-detection algorithms, comparing before
and remote sensing data to identify areas of change and its respective scale, which is then
correlated to damage. For manual processes, innovative approaches have been developed to
crowdsource the analysis of large quantities of images rapidly. Crowd-sourced imagery
analysis was conducted following Hurricane Katrina in 2005, the 2008 Sichuan earthquake
and other disasters. However, it was following the 2010 earthquake in Haiti, when a group of
over 600 volunteers were trained and tasked to identify destroyed and heavily damaged
buildings by the Global Earth Observation-Catastrophe Assessment Network (GEO-CAN),
that the potential for crowdsourcing damage assessment was truly recognized (Ghosh et al.
2012). Since then, similar initiatives have been conducted following major disasters by
organizations such as the non-profit Humanitarian OpenStreetMap Team (HOT) and Digital
Globe’s Tomnod program (Westrope et al. 2014; Barrington et al. 2012).
Remote-sensing based damage assessments can play an important role in providing data
during the period between one week to a few months following a disaster. Since they are
based on direct observation, they can uncover subtle patterns of spatial distribution that may
not emerge from modeled damage estimates. For instance, they may reveal areas with
disproportionate damage due to local ground-motion amplification caused by soil conditions
or other phenomena that cannot be accounted in modeled estimates. This is especially true in
regions with poor geologic, seismologic, exposure, and/or vulnerability data. One of their
more promising uses is in support of the PDNA. As previously discussed, data gathering and
analysis for the PDNA is often an ad-hoc process relying on secondary data, subjective
interpretation and extrapolations. Remote-sensing techniques could theoretically provide
damage information consistent across geography and time, with well-documented methods
and assumptions, to complement standard data used in PDNAs. To facilitate this activity, the
roster of PDNA specialists within the international development community can be trained
up-front to interpret and use remote-sensing based damage assessment results.
Despite their potential, remote-sensing based damage assessments have so far had little
traction in the PDNA process. Concerns about the accuracy of such assessments have proven
particularly difficult to surmount. Past studies have shown significant error in building
damage estimation from both manual and automated remote-sensing based estimates (Booth
et al. 2011; Ghosh et al. 2012; Westrope et al. 2014; Corbane et al. 2011; Brunner et al.
2010). However, it should be noted that these studies assessed accuracy related to individual
building-level damage. It is likely that remote-sensing based damage assessments, when
aggregated to district or regional scales, could provide statistically accurate results well
suited for the needs of the PDNA.
In Nepal, a variety of efforts at remote damage assessments took place within the first 2-3
weeks of the event. Cloud-cover over much of the affected area in the first few days after the
earthquake delayed some of this work, but post-event imagery became available toward the
end of the first week of the response. Unmanned aerial vehicle (UAV) pilots ran afoul of the
Nepali authorities, who banned their use in the first week following the disaster, so this
technology was not used in significant ways for damage assessment. We describe several of
the more prominent activities in the area of remote-sensing for damage assessment during the
Nepal earthquake response.
Within 72 hours of the earthquake, the United States National Geospatial-Intelligence
Agency (NGA) launched a Nepal Earthquake Open Data portal
( that hosted a number of spatial datasets relevant to
the response, including a remote damage assessment initially created on April 28 but updated
through May 13 as new imagery became available. The dataset contains 9317 damaged
structures, categorized as Destroyed (4805), Major (2661), Minor (1375), and Affected (690).
The site contained no information or meta-data to guide users on the ways in which this scale
was applied or should be interpreted. In addition, because undamaged facilities were not
tagged as unaffected, it is difficult to assess both the completeness of the assessment as well
as the relative severity of overall damage in areas of varying building density.
Copernicus is a European Union system that combines remote sensing observations with
"in situ" data collection through sensors and other means. With technical support from the
Information Technology for Humanitarian Assistance, Cooperation and Action (ITHACA),
Copernicus created damage assessments called "grading map", and distributed them through
their data portal ( Building damage classification assessed
"points of interest" such as schools and health facilities (from the open-source GeoNames
database), as well as transportation and utilities, but did not assess individual homes.
Digital Globe's Tomnod initiative is a crowd-sourcing application that uniquely combines
gamified volunteer participation, multi-pass and expert assessment, and algorithmic veracity
scoring for damage estimation and other applications. Over 58,000 volunteers participated in
marking 21,000 damaged structures between April 27th and May 29th. Users were asked to
mark structures as affected or classify areas of major destruction where significant damage
makes it impossible to identify individual buildings. The resulting data was made available
through the project website. Digital Globe released significant amounts of satellite imagery
under permissive licensing at the time as well. Due to the proprietary nature of Tomnod's
software, it is difficult to assess the effectiveness of the veracity scoring as a means of quality
control. Other crowd-sourced efforts, such as those by Kathmandu Living Labs (KLL) and
the Humanitarian OpenStreetMap Team (HOT) were very important for post-earthquake
mapping activities, but they did not undertake building damage estimation at scale.
In addition, several automated remote-sensing based damage assessments were
conducted. In particular, the Advanced Rapid Imaging and Analysis (ARIA) team at NASA’s
Jet Propulsion Laboratory (JPL) and Caltech developed Damage Proxy Maps for the
Kathmandu valley, using change-detection algorithms on interferometric synthetic aperture
radar data collected by the Japanese Space Agency (Yun et al. 2015). These damage maps
correlated well with damage identified by NGA based on optical imagery, and have the
advantage of being insensitive to cloud cover. However, the algorithms currently only
provide information on the spatial distribution and relative scale of damage, but no
information on the number of buildings affected, estimated fatalities or estimated losses.
Another initiative by the NASA/JPL team and ImageCat combined both prediction
modeling and remote sensing to provide estimates of total buildings damaged, losses,
fatalities and their spatial distribution (Huyck 2015). While this assessment had significant
uncertainties (e.g. estimated number of building damage ranged from 225,000 to 450,000), it
was delivered quickly (less than a month following the earthquake) and in a form that
matched user-needs well. Such mixed models are very promising.
A critical early response activity following earthquakes is the evaluation of building
safety for re-occupation. One of the most important drivers of continued risk following an
earthquake is heavy damage to buildings, which can cause additional fatalities and injuries
during aftershocks, under severe weather conditions, or simply due to fatigue. At the same
time, buildings that are damaged but safe to occupy need to be re-opened as soon as possible,
in particular those providing critical services. Broadly speaking, the purposes of engineering
building safety assessments are: (1) to protect human life from the potential dangers of
occupying unsafe buildings, (2) to minimize homelessness and loss of economic activity by
quickly identifying buildings that are safe to occupy, (3) to identify causes of building
damage, to inform improved building and construction standards, and (4) to guide recovery
planning, estimating funding needs, and allocating available resources.
Numerous methodologies have been developed for rapid visual post-disaster safety
assessment of buildings (ATC-20, Japan Building Disaster Prevention Assessment, etc.). The
most commonly used and referenced is the ATC-20 rapid assessment methodology (Applied
Technology Council 1989). It consists of a rapid visual inspection of the exterior of a
building in order to qualitatively assess a building’s residual gravity and lateral force-
resisting capacity, resulting in a building safety placard (green, yellow or red) posted on the
building to indicate building occupancy limitations. For complex buildings or buildings with
uncertain residual capacity, a detailed assessment can be initiated. It should be noted that for
most affected households, the visit from the building safety inspection team is their first
interaction with an organized recovery effort. As such, the safety assessment can serve as a
unique opportunity to inform the public of recovery plans, quell rumors, share information on
safe construction practices, guide affected people towards recovery resources, and more.
In the immediate aftermath of the earthquake, several rapid visual building safety
assessments were launched. The methodology for the assessment is described in “Seismic
Vulnerability Evaluation Guideline for Private and Public Buildings - Part II: Post Disaster
Damage Assessment” (National Society for Earthquake Technology - Nepal 2009). The
methodology originates from various documents produced by FEMA and ATC. Buildings are
assessed through visual inspection and categorized as: inspected with no apparent safety
issues (green placard), limited occupancy (yellow placard), and unsafe (red placard).
With support from the Department of Urban Development and Building Construction
(DUDBC), public-sector buildings were assessed by their respective line-ministries through
the national or district offices. The Nepali Association of Engineering (NEA) coordinated
rapid visual building safety assessments for private buildings. Within two weeks of the
earthquake, the NEA mobilized 3,000 volunteer engineers and architects trained by NSET on
rapid visual building assessment. In the month following the earthquake, these volunteers
were organized into 300 mobile evaluation teams and assessed over 60,000 buildings.
Most evaluations were in direct response to calls from building owners requesting
evaluation of their buildings. These evaluations protected the safety of building tenants and
provided important reassurance to people traumatized by the earthquake and its recurring
aftershocks. However, the assessment faced many challenges and limitations. Despite
numerous calls for funding support, the assessment was never properly funded. It therefore
relied almost entirely on volunteer evaluators, and on donations for vehicles and gas, and
most of the assessments were on paper forms, making it difficult to consolidate results into a
single database. In addition, because the assessment was coordinated from the capital city of
Kathmandu, relatively few buildings were evaluated outside of Kathmandu valley, including
those districts most affected by the earthquake. Finally, since it was not a systematic door-
to-door assessment, the results were not suitable for use as a data-base for recovery planning.
The rapid visual building inspections described in the previous section have as their main
purpose the identification of potential dangers related to damage caused by the earthquake.
They do not provide information on what to do about said damage. As such, there is a need
for a recovery-oriented damage assessment in addition to the previously described safety-
oriented building damage assessment to assist national governments to plan, budgetize,
implement and monitor recovery activities. We focus here on the recovery-oriented damage
assessment conducted in Nepal in support of owner-driven reconstruction programs. Many
studies have demonstrated the effectiveness of the Owner-Driven Reconstruction (ODR)
approach (Jha et al. 2009; Barenstein 2008). Its most salient aspect is that households are
empowered to control their own recovery through provision of reconstruction grants over
several installments, with each new transfer of funds triggered by an inspection verifying
adherence to safe construction practices. ODR programs are accompanied by significant
community mobilization, construction training, and close management and scale-up of
construction material supply chains (Arshad & Athar 2013). In the following section, we use
the example of Nepal’s Earthquake and Household Damage and Characteristics (EHDC)
survey as an example of recovery-oriented damage surveys, and to demonstrate the role of
this survey in the promotion of effective recovery strategies.
The Government of Nepal stated in the PDNA that the recovery of the housing sector “would
be based on the principles of equity, inclusion and community participation through an
owner-driven reconstruction (ODR) approach to build back better” (Government of Nepal
2015). In anticipation of this, the World Bank prepared a support program for a housing
recovery project, to be enabled by a comprehensive damage and household vulnerability
survey (The World Bank 2015). Even though the World Bank support program did not
finance the full needs of the housing recovery, it was necessary to design and implement a
survey that would ensure uniformity in the beneficiary selection process across the sector.
Although the original plan for the damage and beneficiary eligibility survey was not fully
implemented, it is worth looking at the original objectives (Lallemant 2015). These were:
1. To record the complete scope of damages to the housing stock;
2. To develop the list of beneficiaries to receive housing recovery assistance, in a manner
that is uniform, equitable and inclusive;
3. To provide on-the-spot bank accounts for affected households, so as to most quickly
transfer recovery assistance to support owner-driven recovery efforts;
4. To determine preliminary site safety that could trigger relocation requirement at the
household level (this would be followed up with more in-depth site assessment);
5. To set up and populate a management information system (MIS) to coordinate, monitor
and identify obstacles to recovery activities;
6. To promote short-term and long-term risk reduction through communication of safe
construction messages by evaluation teams;
7. To promote public reassurance that recovery is underway;
8. To support the implementation of an installment-based housing recovery assistance
The original design of the program was focused on catalyzing the housing recovery
process, and to do so to higher seismic standards. Since the level of reconstruction assistance
would be based in large part on the level of damage to the home, the assessment would serve
as a key criterion for beneficiary enrollment in the recovery program. The damage
assessment methodology itself would be action-oriented, such that the level of damage could
be matched to specific recovery actions (repair, reconstruction, or strengthening/retrofitting)
and linked to corresponding financial assistance packages. Homeowners would be enrolled
on the spot in the recovery program by signing a Memorandum of Agreement (MoU) with
the government inspector, acknowledging the assessment results and agreeing to follow
prescribed safe construction practices. Assistance would be provided in installments, with
inspections made before the release of each. Since few of the affected people had access to
banking services and even fewer had bank accounts, the assessment would also be a
mechanism to open accounts on the spot for beneficiaries of reconstruction assistance. Banks
could be convinced to waive or greatly simplify enrollment criteria for accounts, especially
since this would be an opportunity for them to greatly expand their clientele.
The intention was also to use the assessment as a mechanism to shift the construction
culture towards seismically resilient practices. Inspectors would communicate information on
common seismic design flaws, coupled with information on safe construction practices.
Community facilitators would accompany the assessment teams in order to address
grievances locally as much as possible, and answer questions on the recovery program more
broadly. Finally, all damage data, geographic coordinates, pictures, MoUs and grievances
would be recorded on digital tablets linked to a central Management Information System
(MIS). This MIS would serve not only as a data repository and beneficiary list, but also as a
platform to coordinate and monitor reconstruction activities, as well as identify bottlenecks,
delays or other obstacles to recovery. This recovery-oriented survey design was not
implemented in its entirety, but nonetheless formed the seed for the Earthquake and
Household Damage and Characteristics (EHDC) survey that was eventually put together.
It was only on December 25, 2015 that the National Reconstruction Authority (NRA) was
established, eight months after the earthquake. Prior to its establishment, leadership remained
unclear and decision-making regarding critical details of EHRP, including eligibility
requirements and the EHDC Survey, were not settled. The EHDC Survey scheduled for
completion by September 2015, but the final project agreement was not finalized between
UNOPS and the Central Bureau of Statistics (CBS) until October 2015. Delays were
exacerbated by political unrest due to the passage of a new constitution in September 2015
and a change of government in October 2015. The political unrest led to frequent closures of
Nepal’s dry ports with India, rendering petrol difficult to acquire in the volumes needed to
transport surveyor teams to visit each of the approximately 715,000 houses planned for in the
first phase of the survey (The World Bank 2016). In December 2015, it was decided by the
Steering Committee to proceed. Houses were already being reconstructed and the fear was
that if not done quickly, the survey would no longer be of use.
Accordingly, the first phase of the survey, covering the 11 heavily affected districts,
began in January 2016 and was mostly completed by May 2016. It was later decided to also
survey the three Kathmandu Valley districts using a refined methodology, a process which
was mostly completed by end-September 2016 (National Reconstruction Authority - Nepal
2016). Implementation of the EHDC Survey required the hiring, training, and deployment of
over 1,500 staff (UNOPS 2016). Surveyor teams consisted of one engineer and one social
mobilizer equipped with tablets for recording collected attributes. SIM cards with data were
used to upload the data to the Government Integrated Data Center (GIDC), from which the
data was cleaned and uploaded to the MIS by the Central Bureau of Statistics (CBS).
Eligibility requirements for beneficiaries were determined by the National Reconstruction
Authority (NRA) after the survey began.
The authors have collectively spent nearly two years following and monitoring the
deployment and implementation of the EHDC survey and the EHRP. The following lessons
learned are based on personal experiences and meetings with stakeholders of the survey and
reconstruction program. The intent is to serve as guide for the development of future
recovery-oriented damage surveys.
(1) The need to clarify the type, scope and purpose of each damage assessment initiative. The
multitudes of damage assessments conducted following disaster creates significant confusion.
This paper attempts to bring some clarity and common descriptive language surrounding such
initiatives. Damage assessments are only relevant in their relation to the response / recovery
activities that they enable. As such, this needs to be defined and communicated up-front, so
as to minimize confusion and conflict, and reduce the potential for misuse. This issue is
exemplified by the early damage survey conducted by the Ministry of Home Affairs in the
weeks following the earthquake. The non-technical rapid assessment led to the distribution of
Earthquake Victim ID Cards (EVCs). This early assessment was not transparent about
assessment criteria, the meaning of EVC cards or what the results would be used for. This led
to delays, complications and conflict as the disaster shifted from response to recovery, in
particular since holding an EVC had no impact on eligibility for individual housing
reconstruction grants under EHRP, which was determined from the EHDC Survey.
(2) The need for extensive and transparent quality control and grievance mechanisms. The
integrity of the survey hinges upon the accuracy and consistency of assessments across
survey teams, the effective communications of assessment criteria and the ability to address
grievances quickly and clearly. If the surveyed data is undermined by even a few “bad”
assessments, the credibility of the entire program will be at risk. The rapid mobilization of
manpower presents a risk that survey teams will require more training to perform satisfactory
inspections. Although damage surveys are inherently time-sensitive, this aspect should not be
compromised and quality control mechanisms should be planned and implemented alongside
the survey. In Nepal, the EHRP dealt with many of these issues after the survey began, and
has faced media scrutiny regarding survey omissions. EHRP also recorded a large number of
grievance cases from individuals believing that the survey results were not correct (National
Reconstruction Authority - Nepal 2016). These issues are time and resource intensive to
address, and provide easy targets for program critics.
(3) The need to take full advantage of the logistical effort of the survey. The logistical task of
sending survey teams to each affected building is incredible, especially in disasters with large
geographic coverage and hard-to-access communities. As such, the survey should be seen as
a unique opportunity to share information to affected households (e.g. about the
reconstruction program, about safe construction guidelines, about disaster risk awareness,
etc), and an opportunity to jump-start other aspects of the reconstruction program (e.g.
opening up bank accounts for beneficiaries, identifying buildings which could be used for
local reconstruction training or community gathering, identifying communities with
particular vulnerabilities or needs, etc.). Such piggyback activities require additional up-front
planning, staffing and resources, but have significant multiplicative impacts on recovery. In
Nepal, the EHDC Survey did not collect all information necessary to open bank accounts,
requiring a second level of engagement with eligible beneficiaries. Nor did EHDC Survey
communicate details of the program, wasting a valuable opportunity to inform and engage
with each potential beneficiary directly and resolve certain grievances on the spot. Nor did
the EHDC serve to automatically enroll beneficiaries in the program, requiring instead a
separate beneficiary enrollment survey which delayed 1st installment disbursements by as
much as six month, and incurring additional millions in cost.
This paper presents the primary building damage assessments conducted in the aftermath of
the 2015 Gorkha earthquake in Nepal. Each type of damage assessment is described,
including its purpose, general methodology, time-frame, and specific implementation in the
context of the Nepal earthquake. In order to clarify the language surrounding these tools, the
paper serves to describe a typology and taxonomy of post-earthquake damage assessment.
The authors propose five typologies of damage assessments that have become standard in
post-earthquake environments: (1) modeled disaster damage estimates, (2) the Post-Disaster
Needs Assessment (PDNA), (3) remote-sensing based damage assessments, (4) rapid visual
engineering building safety assessments and (5) recovery-oriented damage surveys. These are
distinctive in methodology, time-frame and purpose, underlining the need for specific
terminology beyond generic references to “damage assessments.”
Particular focus was placed on recovery-oriented damage surveys as a key mechanism
through which various recovery activities can take shape: identification of beneficiaries,
detailed allocation of resources, communication on recovery plans and safe construction
practices and more. It was found that Nepal’s recovery-oriented damage survey was
instrumental in developing the beneficiary list for the Earthquake Housing Reconstruction
Program, though it faced numerous delays and complications. Key lessons learned and
recommendations are provided. Overall, these reinforce the need for early planning and
decision-making in support of comprehensive recovery-oriented housing damage surveys, so
as to promote rapid, transparent and resilient post-earthquake housing recovery.
This work was partially supported by the National Science Foundation (NSF) Grant No
106756, the NSF Graduate Research Fellowship Program, the John Blume Earthquake
Engineering Research Center and the Stanford School of Engineering Graduate Fellowship.
We also thank the Government of Nepal, the Nepal PDNA steering committee, GFDRR, the
World Bank, NSET and USGS’s PAGER team for their support and advice.
Applied Technology Council, 1989. Procedures for Postearthquake Safety Evaluation of Buildings.
Arshad, S. & Athar, S., 2013. Rural housing reconstruction program post-2005 earthquake: Learning
from the Pakistan Experience, World Bank Publication.
Barenstein, J.D., 2008. From Gujarat to Tamil Nadu: Owner-driven vs. contractor-driven housing
reconstruction in India, Proceedings of the 4th International i-Rec Conference.
Barrington, L. et al., 2012. Crowdsourcing earthquake damage assessment using remote sensing
imagery. Annals of Geophysics, 54(6).
Binns, C., 2016. How do you recover after a deadly earthquake? Available at:
Booth, E. et al., 2011. Validating Assessments of Seismic Damage Made from Remote Sensing.
Earthquake Spectra, 27(S1), pp.S157S177.
Brunner, D., Lemoine, G. & Bruzzone, L., 2010. Earthquake Damage Assessment of Buildings Using
VHR Optical and SAR Imagery. IEEE Transactions on Geoscience and Remote Sensing, 48(5).
Corbane, C. et al., 2011. A Comprehensive Analysis of Building Damage in the 12 January 2010
Mw7 Haiti Earthquake Using High-Resolution Satellite and Aerial Imagery. Photogrammetric
Engineering Remote Sensing, 77(10), pp.09971009.
Dong, L. & Shan, J., 2013. A comprehensive review of earthquake-induced building damage
detection with remote sensing techniques. ISPRS Journal of Photogrammetry and Remote
Sensing, 84, pp.8599.
Earle, P.S. et al., 2009. Prompt Assessment of Global Earthquakes for Response (PAGER): A system
for rapidly determining the impact of earthquakes worldwide.
GFDRR, EU, UNDP, 2013. Post-Disaster Needs Assessments Guidelines - Volume A.
Ghosh, S. et al., 2011. Crowdsourcing for Rapid Damage Assessment: The Global Earth Observation
Catastrophe Assessment Network (GEO-CAN), Earthquake Spectra, 27(S1) pp.S179-S198.
Gnyawali, K.R. & Adhikari, B.R., 2016. Spatial Relations of Earthquake Induced Landslides
triggered by 25 April 2015 Gorkha Earthquake, Mw 7.8, Nepal. EGU General Assembly
Conference Abstracts, 18, p.18429.
Government of Nepal, 2015. Nepal Earthquake 2015 Post Disaster Needs Assessment - Vol. A: Key
Findings, Kathmandu.
Huyck, C., Gorkha-NASA/JPL&ImageCat Assessment
Jaiswal, K. & Wald, D.J., 2013. Strategies for rapid global earthquake impact estimation: the Prompt
Assessment of Global Earthquakes for Response (PAGER) system.
Jha, A.K. et al., 2009. Safer Homes, Stronger Communities. pp.1407.
Joyce, K.E. et al., 2009. A review of the status of satellite remote sensing and image processing
techniques for mapping natural hazards and disasters. Progress in Physical Geography, 33(2).
Lallemant, D., 2015. Personal Notes on Housing Damage Survey - Kathmandu World Bank Offices.
National Reconstruction Authority - Nepal, 2016. Rebuilding Nepal, Kathmandu.
National Society for Earthquake Technology - Nepal, 2009. Seismic Vulnerability Evaluation
Guideline for Private and Public Buildings - Part II: Post Disaster Damage Assessment,
The World Bank, 2016. Earthquake Housing Reconstruction Project (P155969),
The World Bank, 2015. Earthquake Housing Reconstruction Project - World Bank PAD pp.170.
UNOPS, 2016. Rebuilding lives and homes in Nepal, Available at:
Voigt, S. et al., 2007. Satellite Image Analysis for Disaster and Crisis-Management Support. IEEE
Transactions on Geoscience and Remote Sensing, 45(6), pp.15201528.
Westrope, C., Banick, R. & Levine, M., 2014. Groundtruthing OpenStreetMap Building Damage
Assessment. Procedia Engineering, 78, pp.2939.
Yun, S.-H. et al., 2015. Rapid Damage Mapping for the 2015 Mw 7.8 Gorkha Earthquake Using
Synthetic Aperture Radar Data from COSMOSkyMed and ALOS-2 Satellites. Seismological
Research Letters, 86(6), pp.15491556.
... Examples of such approaches include the tephra-fall impact assessment by Magill et al. (2013) Compared to other phenomena, the relatively low frequency of impactful eruptions is a limiting factor to the availability of postevent impact assessments-both field and RS surveys-for volcanic hazards. The literature on earthquakes and hurricanes illustrates the use of several RS-based damage assessments methods including (i) manual visual inspections of high to very high-resolution satellite and aerial optical imagery (Corbane et al., 2011), (ii) automated change-detection algorithms on synthetic aperture radar (SAR) data (Yun et al., 2015), and (iii) a combination of RS data with prediction modeling (Lallemant et al., 2017). As of yet, automated damage assessments from optical imagery, either satellite or aerial, has had limited success due to the complexity of the task and the lack of context-specific data to train the algorithms. ...
... As of yet, automated damage assessments from optical imagery, either satellite or aerial, has had limited success due to the complexity of the task and the lack of context-specific data to train the algorithms. As a result, most methods still rely on manual damage identification, which has led to the emergence of volunteer crowd-sourcing initiatives to manually inspect imagery, often covering hundreds of thousands of potentially affected buildings and infrastructure (Ghosh et al., 2011;Lallemant et al., 2017). Various volunteer technical communities and platforms have been developed to support this manual imagery inspection task, including the Humanitarian Openstreetmap Team (HOT; ...
In volcanology, remote sensing provides a global observation framework that is becoming increasingly valuable at multiple spatio-temporal scales. Satellite observations have improved monitoring for even the most remote volcanoes, provided a large-scale context to extrapolate field-based observations, and is now routinely used to characterize and monitor volcanic systems and their hazards. However, remote sensing remains underutilized to characterize volcanic impacts that result from the interaction of volcanic processes and societal assets. This chapter first provides a review of applications of remote sensing for impact assessment in volcanology, both in pre- and post-event contexts. Two case studies are used to illustrate the application of recent remote-sensing methods in impact assessments within the field of volcanology. First, the impact to the built environment caused by the 2014 eruption of Kelud volcano (Indonesia) is assessed using automated damage proxy maps produced from interferometric synthetic-aperture radar data. This automated approach is compared, validated and discussed against an impact assessment performed from the manual inspection of high-resolution satellite images. Second, Google Earth Engine is used to illustrate how simplified access to remote-sensing data offered by cloud-based platforms can provide a global and systematic framework to revisit the impacts of eruptions We illustrate the application of Google Earth Engine to various tasks, such as change detection and time series analyses, using the 2010 eruption of Merapi volcano (Indonesia). Very few volcanic impact assessments fully utilize remote sensing, but as such data become more easily available, stored and processed, there is the exciting potential for new methods and tools that can better capture, characterize and ultimately forecast volcanic impacts.
... Examples of such approaches include the tephra-fall impact assessment by Magill et al. (2013) Compared to other phenomena, the relatively low frequency of impactful eruptions is a limiting factor to the availability of postevent impact assessments-both field and RS surveys-for volcanic hazards. The literature on earthquakes and hurricanes illustrates the use of several RS-based damage assessments methods including (i) manual visual inspections of high to very high-resolution satellite and aerial optical imagery (Corbane et al., 2011), (ii) automated change-detection algorithms on synthetic aperture radar (SAR) data (Yun et al., 2015), and (iii) a combination of RS data with prediction modeling (Lallemant et al., 2017). As of yet, automated damage assessments from optical imagery, either satellite or aerial, has had limited success due to the complexity of the task and the lack of context-specific data to train the algorithms. ...
... As of yet, automated damage assessments from optical imagery, either satellite or aerial, has had limited success due to the complexity of the task and the lack of context-specific data to train the algorithms. As a result, most methods still rely on manual damage identification, which has led to the emergence of volunteer crowd-sourcing initiatives to manually inspect imagery, often covering hundreds of thousands of potentially affected buildings and infrastructure (Ghosh et al., 2011;Lallemant et al., 2017). Various volunteer technical communities and platforms have been developed to support this manual imagery inspection task, including the Humanitarian Openstreetmap Team (HOT; ...
The ensemble of pyroclastic material injected into the atmosphere during explosive volcanic eruptions (i.e., tephra) has the potential to generate complex patterns of disruption at multiple temporal and spatial scales. Reducing associated risk requires a full characterization of the hazards associated with its dispersal, sedimentation, and aeolian remobilization and of various dimensions of vulnerability (e.g., physical, socio-economic, and systemic). Dedicated analytical and numerical models together with sophisticated probabilistic strategies have been developed for the description of the physical processes; however, due to the combination of a low frequency of volcanic eruptions and the infancy of associated risk studies, the characterization of vulnerability is still fragmentary and mostly focused on the physical acute damage. In this context, impact analyses have already provided fundamental insights into the vulnerability of various elements and systems and are essential to identifying important drivers of losses. Strategies of disaster risk reduction in volcanic regions will be effective only after the potential dynamic interaction between the geophysical event and exposed elements are understood and described.
... In relation to the subject of this article, in [1], [2] we find several prevention proposals to a) mitigate damage in seismic zones, b) reduce risks to the population and c) reduce material damage. In [3], [4], they point out the effectiveness of protocols to reduce losses and indirect costs such as interruption of productive activities and basic services and the effects of post-traumatic stress. Other works [5], [6], [7], emphasize the complexity of disaster scenarios and the need for the different actors involved in damage repair to share unique information, in particular the use of mobile computing and social networks. ...
On September 19, 2017 an earthquake occurred in Mexico with epicenter in the limits of the states of Puebla and Morelos. It had a magnitude of 7.1 on the Richter scale. 31,090 homes were affected, of which 1659 with total damage. In order to attend to the contingency, the government of Morelos formed an inter-institutional committee in the first hours to carry out an initial diagnosis of the damage and to provide emergency services. This article presents a case study and lessons learned from the software engineering support for the development of a data-driven platform in the various phases of contingency response: census of damaged homes, identification of aid beneficiaries, determination of aid packages according to a damage assessment, logistics and follow-up of aid package delivery, data-driven decision making, and a public portal for open data and budget transparency. KEYWORDS
An earthquake of moment magnitude (Mw) 7.8, the largest in the last 80 years, struck central Nepal on April 25, 2015. Named the Gorkha earthquake, it wreaked havoc on the country’s central region, affecting 32 of 77 districts. The earthquake impacted nearly a million private houses, thousands of educational infrastructures, hundreds of health facilities, myriad cultural heritages and many other infrastructures. The National Reconstruction Authority (NRA) was established to lead the post-earthquake recovery and reconstruction and complete it within five years. This paper presents details of the earthquake’s effects, the early response and the status of the post-earthquake reconstruction progress five years after the earthquake. Official reports from Nepalese government institutions, national and international authorities, government databases and research papers on the Gorkha earthquake were reviewed. The numbers of various infrastructures reconstructed and under construction each year after the earthquake and the reconstruction processes adopted are presented. The results show that 67% of private houses, 74% of educational institutions and 58% of health facilities were reconstructed in the five years.
The ecological environment damage caused by human activities will aggravate the damage of natural disasters and seriously restrict the socioeconomic sustainable development. Therefore, it is a key issue to coordinate the coupling relationship between socioeconomic development (SE) and ecological environment (EE) to reduce the damage of natural disasters to human activities and guarantee the socioeconomic sustainable development after natural disasters. It is an important issue to explore the relationship between economy and environment in post-disaster reconstruction areas and how to coordinate the areas with slower economic development and higher risk of ecological destruction. Based on this, an index system of ecological environment and socioeconomic development was constructed by using the data in the statistical yearbook from 2005 to 2018. The dynamic deviation maximization (DM) method and coupling coordination degree model (CCDM) were used to study the coordination relationship between socioeconomic and ecological environment in Wenchuan earthquake disaster area. The results showed that the coupling coordination degree (CCD) presented an upward trend in the period of post-disaster reconstruction in Wenchuan earthquake disaster area, and the CCD is highly related to the regional topography, city location, urban comprehensive economic level, and urban ecological endowment. The results indicated that the CCD shows a downward trend from provincial capitals to surrounding cities on the basis of the perspective of urban spatial distribution characteristics. On the basis of the research, this paper analyzes the reasons for the differences in CCD scores of different cities and puts forward corresponding policy suggestions, which including speeding up industrial transformation, increasing government investment, and strengthening transportation infrastructure construction. By analyzing the CCD between SE and EE in Wenchuan earthquake disaster area, this paper aims to provide reference for achieving regional sustainable development goals (SDGs) in post-disaster reconstruction, as well as hope to provide an experience reference for post-disaster reconstruction in other disaster areas.
Full-text available
The increasing frequency and severity of wildfire events in the last few decades has created an urgent need for new technologies that allow rapid surveying and assessment of post-wildfire building damage. However, existing technologies lack in accuracy and ability to scale to effectively aid disaster relief and recovery. Even today, most wildfire event inspectors need to physically visit the areas impacted by wildfires and manually classify building damage, requiring considerable time and resources. Here, we present DamageMap, an artificial intelligence-powered post-wildfire building damage classifier. DamageMap is a binary classifier (outputs are –“damaged” or “undamaged”). Unlike existing solutions that require both pre- and post-wildfire imagery to classify building damage, DamageMap relies on post-wildfire images alone by separating the segmentation and classification tasks. Our model has an overall accuracy of 98% on the validation set (five wildfire events all around the world) and 92% and 98% on two independent test sets from the Camp Fire and the Carr Fire, respectively. Excellent model performance across a variety of datasets provides evidence of DamageMap's robustness to unseen data. Thus, DamageMap may help governmental and non-governmental agencies rapidly survey building damage using post-wildfire aerial or satellite imagery in wildfire-impacted areas. DamageMap is available as a server-side web-application.
Full-text available
Probabilistic loss assessments from natural hazards require the quantification of structural vulnerability. Building damage data can be used to estimate fragility curves to obtain realistic descriptions of the relationship between a hazard intensity measure and the probability of exceeding certain damage grades. Fragility curves based on the lognormal cumulative distribution function are popular because of their empirical performance as well as theoretical properties. When we are interested in estimating exceedance probabilities for multiple damage grades, these are usually derived per damage grade via separate probit regressions. However, they can also be obtained simultaneously through an ordinal model which treats the damage grades as ordered and related instead of nominal and distinct. When we use nominal models, a collapse fragility curve is constructed by treating data of “near-collapse” and “no damage” the same: as data of noncollapse. This leads to a loss of information. Using synthetic data as well as real-life data from the 2015 Nepal earthquake, we provide one of the first formal demonstrations of multiple advantages of the ordinal model over the nominal approach. We show that modeling the ordering of damage grades explicitly through an ordinal model leads to higher sensitivity to the data, parsimony and a lower risk of overfitting, noncrossing fragility curves, and lower associated uncertainty.
After an earthquake, many responding organizations need to understand the scale and distribution of building damage to react effectively. However, their building damage information needs and information use remain poorly understood, limiting the efficacy of information production, sharing, and research. To clarify those needs, we conducted a two-part survey, comprising semi-structured interviews and an online questionnaire, of building damage information users and providers. Based on the interview data and questionnaire responses, we characterize six post-disaster tasks that rely on building damage information by their timing and by the necessary qualities of the information they require. Through inductive analysis of the interview data, we show that responders’ use of building damage information also depends on factors beyond the building damage information itself—namely, trust, impediments to information sharing, their varying understandings of disaster, and their attitudes toward emerging technologies. These factors must be considered in the design of any effort to create and/or disseminate post-disaster building damage information.
Full-text available
Earthquake affected households too often insufficiently apply seismic construction knowledge during reconstruction. This study aims to assess to what degree safety guidelines have found their way to practice in Nepal. Differences are explored between communities in the Gorkha and Okhaldhunga districts, which received differing levels of technical assistance following the 2015 earthquakes. Seismic resistance of houses was assessed 3 years after the earthquakes. Findings from 955 houses in 25 communities show high degrees of adoption of earthquake-resistant construction knowledge in all selected communities. Variation in safer construction across communities differs only slightly for different intensities of humanitarian technical assistance. This finding points toward the need to more closely examine the communication methods employed and motivations of households to build back safer.
Purpose Decision-makers, practitioners and community members have a need to assess the disaster resilience of their communities and to understand their own capacities in disaster situations. There is a lack of consensus among researchers as to what resilience means and how it can be measured. This paper proposes a novel technique to achieve consensus among stakeholders on definitions, objectives and indicators for measuring a key dimension of community disaster resilience (CDR), physical infrastructure (PI). Design/methodology/approach This study uses a five-step approach utilizing Q-methods to contextualize a resilience index for PI. Interviews, focus groups and Q-sorting workshops were conducted to develop a tool that ranked measures according to stakeholder preference. A total of 84 participants took part in the workshops across four countries (United Kingdom, Malaysia, Pakistan and Sri Lanka). Findings The initial set of 317 measures was reduced to 128 and divided into the three community capacities of anticipatory, absorptive and restorative. The physical infrastructure capacity assessment tool (PI-CAT) was then finalized to have 38 indicators that were also ranked in order of importance by the participants. Practical implications The PI-CAT can be useful for local governments and communities to measure their own resilience. The tool allows stakeholders to be confident that the metrics being used are ones that are relevant, important and meet their requirements. Originality/value The Q-method approach helps stakeholders to develop and use a community capacity assessment tool that is appropriate for their context. The PI-CAT can be used to identify effective investments that will enhance CDR.
Full-text available
This paper presents the results of an extensive mapping of co-seismic landslides triggered by the 2015 Gorkha earthquake in central Nepal. More than 19,332 landslides have been identified covering 61.5 km2 of land in about 20,500 km2 area of investigation using Google Earth imagery. Their spatial distribution characteristics and relation to the triggering mechanism is studied. Interesting regional localization and angular distribution characteristics, more controlled by the rupture directivity is observed. Seismic, geomorphic and lithological parameters that induce susceptibility to their occurrence is studied using two indices of landslide concentration: Landslide Area Percentage (LAP) and Landslide Number Percentage (LNP) in comparison with % area of each parameter classes. Positive correlation with the chosen triggering parameters are observed but there are some significant differences in the parameter values and distribution plots to co-seismic landslides in other parts of world. These results provide valuable information about the slope response characteristics in case of seismic activation in thrust faulting Himalayan landscapes, and this is important in further researches on co-seismic landslide prediction models for mountainous settlements, sediment yield studies and cascading landslide disasters after major earthquakes.These results provide valuable information about the slope response characteristics in case of seismic activation in thrust faulting Himalayan landscapes, and this is important in further researches on co-seismic landslide prediction models for mountainous settlements, sediment yield studies and cascading landslide disasters after major earthquakes.
Full-text available
The 25 April 2015 Mw 7.8 Gorkha earthquake caused more than 8000 fatalities and widespread building damage in central Nepal. The Italian Space Agency's COSMO-SkyMed Synthetic Aperture Radar (SAR) satellite acquired data over Kathmandu area four days after the earthquake and the Japan Aerospace Exploration Agency's Advanced Land Observing Satellite-2 SAR satellite for larger area nine days after the mainshock. We used these radar observations and rapidly produced damage proxy maps (DPMs) derived from temporal changes in Interferometric SAR coherence. Our DPMs were qualitatively validated through comparison with independent damage analyses by the National Geospatial-Intelligence Agency and the United Nations Institute for Training and Research's United Nations Operational Satellite Applications Programme, and based on our own visual inspection of DigitalGlobe'sWorld-View optical pre-versus postevent imagery. Our maps were quickly released to responding agencies and the public, and used for damage assessment, determining inspection/imaging priorities, and reconnaissance fieldwork.
Full-text available
Assessments of damage following the 2010 Haitian earthquake were validated by comparing three datasets. The first, for 107,000 buildings, used vertical aerial images with a 15-25 cm spatial resolution. The second, for 1,241 buildings, used Pictometry images (oblique angle shots with a resolution of about 10 cm taken in four directions by aircraft). The third dataset, for 142 buildings, used ground observations. The ground observations confirmed the tendency of remote sensing to underestimate the proportion of heavily damaged and collapsed buildings, and the difficulty of making remote assessments of moderate or low damage. Bayesian statistics and sample surveys made from Pictometry images and ground observations were used to improve remote damage assessments from vertical images. The possibility of developing standard factors to correct remote assessments is discussed. The field exercise pointed to the need to produce an internationally agreed-upon set of damage definitions, suitable for postdisaster needs assessments as well as for other uses. [DOI:10.1193/1.3632109]
Full-text available
The paper provides an account of how three key relief organizations worked together after the devastating Haiti earthquake to produce the first damage assessment based mainly on the use of remotely-sensed imagery. This assessment was jointly conducted by the World Bank (WB), the United Nations Institute for Training and Research (UNITAR) Operational Satellite Applications Programme (UNOSAT), and the European Commission’s Joint Research Centre (JRC). This paper discusses the data sources used for the assessment, the methodologies employed to evaluate building damage, and a set of independent studies to validate the final damage results. Finally, a vision of the role of remote sensing technologies in future disasters is presented that serves as a road map for methodological improvements.
Full-text available
To gauge the accuracy of the crowd-sourced damage assessments in the Philippines following Typhoon Haiyan, REACH and the American Red Cross conducted a study comparing enumerated field damage assessments with the remote damage assessments conducted by the OpenStreetMap community. The potential utility of remote sensing imagery and rapid GIS-based mapping in humanitarian responses relies on the accuracy of these techniques. Recent studies from other emergencies have questioned the current capacity of these tools to deliver the levels of accuracy needed, but have acknowledged that these levels can be improved with further research, development and standardization for the humanitarian context. This assessment sought to address some of these questions of accuracy by comparing remote damage assessment findings with field-level damage assessments and to identify any differences in accuracy. The assessment also aimed to assess the ability of crowd-sourced platforms to go beyond providing only base data by creating information about building-level damage. The conclusions and recommendations are intended to inform contributors and developers of crowd-source platforms as well as the humanitarian community at large, contributing to a dialogue about the how to capitalize on the present tools and improve the way in which they are used in humanitarian settings.
Within minutes of a significant earthquake anywhere on the globe, the U.S. Geological Survey (USGS) Prompt Assessment of Global Earthquakes for Response (PAGER) system assesses its potential societal impact. PAGER automatically estimates the number of people exposed to severe ground shaking and the shaking intensity at affected cities. Accompanying maps of the epicentral region show the population distribution and estimated ground-shaking intensity. A regionally specific comment describes the inferred vulnerability of the regional building inventory and, when available, lists recent nearby earthquakes and their effects. PAGER’s results are posted on the USGS Earthquake Program Web site (, consolidated in a concise one-page report, and sent in near real-time to emergency responders, government agencies, and the media. Both rapid and accurate results are obtained through manual and automatic updates of PAGER’s content in the hours following significant earthquakes. These updates incorporate the most recent estimates of earthquake location, magnitude, faulting geometry, and first-hand accounts of shaking. PAGER relies on a rich set of earthquake analysis and assessment tools operated by the USGS and contributing Advanced National Seismic System (ANSS) regional networks. A focused research effort is underway to extend PAGER’s near real-time capabilities beyond population exposure to quantitative estimates of fatalities, injuries, and displaced population.
This chapter summarizes the state-of-the-art for rapid earthquake impact estimation. It details the needs and challenges associated with quick estimation of earthquake losses following global earthquakes, and provides a brief literature review of various approaches that have been used in the past. With this background, the chapter introduces the operational earthquake loss estimation system developed by the U.S. Geological Survey (USGS) known as PAGER (for Prompt Assessment of Global Earthquakes for Response). It also details some of the ongoing developments of PAGER's loss estimation models to better supplement the operational empirical models, and to produce value-added web content for a variety of PAGER users.
This paper describes the evolution of recent work on using crowdsourced analysis of remote sensing imagery, particularly high-resolution aerial imagery, to provide rapid, reliable assessments of damage caused by earthquakes and potentially other disasters. The initial effort examined online imagery taken after the 2008 Wenchuan, China, earthquake. A more recent response to the 2010 Haiti earthquake led to the formation of an international consortium: the Global Earth Observation Catastrophe Assessment Network (GEO-CAN). The success of GEO-CAN in contributing to the official damage assessments made by the Government of Haiti, the United Nations, and the World Bank led to further development of a web-based interface. A current initiative in Christchurch, New Zealand, is underway where remote sensing experts are analyzing satellite imagery, geotechnical engineers are marking liquefaction areas, and structural engineers are identifying building damage. The current site includes online training to improve the accuracy of the assessments and make it possible for even novice users to contribute to the crowdsourced solution. The paper discusses lessons learned from these initiatives and presents a way forward for using crowdsourced remote sensing as a tool for rapid assessment of damage caused by natural disasters around the world.
Earthquakes are among the most catastrophic natural disasters to affect mankind. One of the critical problems after an earthquake is building damage assessment. The area, amount, rate, and type of the damage are essential information for rescue, humanitarian and reconstruction operations in the disaster area. Remote sensing techniques play an important role in obtaining building damage information because of their non-contact, low cost, wide field of view, and fast response capacities. Now that more and diverse types of remote sensing data become available, various methods are designed and reported for building damage assessment. This paper provides a comprehensive review of these methods in two categories: multi-temporal techniques that evaluate the changes between the pre- and post-event data and mono-temporal techniques that interpret only the post-event data. Both categories of methods are discussed and evaluated in detail in terms of the type of remote sensing data utilized, including optical, LiDAR and SAR data. Performances of the methods and future efforts are drawn from this extensive evaluation.