Hazard Assessment and Exposure Modeling

Using Digital Data in the Himalayas: An

Experience of Nepal

Amod Mani DIXIT, President, NSET, email: [email protected]

Surya Narayan SHRESTHA, Executive Director, NSET,
email: [email protected]
Ganesh Kumar JIMEE, Deputy Executive Director, NSET,
email: [email protected]
Sharad WAGLE, Structural Engineer, NSET, email: [email protected]

04/10/2023 1

Baliyo Ghar Program

 DesInventar Database
• From 1971 to Today
• 30 Hazard Types, including landslide, earthquake, animal
attack etc
Event Municipali
Date Time State Region District
Cause ty
Evidence
for ongoing
Placenam Longitude GPS Preinform
Ward Latitude (post-
e Location ed failure)
How was risk?
the How was the
How was the Male Female Total
rainfall /
landslide location Injured
described?
trigger
described? Death Death Death
described?

Affected Damage Destroyed Source Source

Misssing
People houses Houses date URL

Baliyo Ghar Program

 LANDSLIDE MAPPING

• 2014-2020 pre- and post-

monsoon landslide
mapping in red grid (5x5
km)
• 2016-2021 post-monsoon
landslide mapping in pink
grid (5x5 km) for training
data
• Expanding the work of 14
earthquake affected
district to all over Nepal by
Semiautomated method

Baliyo Ghar Program

 Semi-automated landslide
Inventory mapping
Expanded from 14 districts to all over Nepal Satellite Image
Training Dataset:
1. Landslides polygons manually digitized
2. For Pre and Post monsoon epochs from 2014 to 20 of
14 Districts (orange area in above map)
3. Post monsoon epochs of 300 grids(purple squares in
above map) 5KM by 5KM grids covering all Nepal of Segmented
Feature Extraction and
years 2016 to 2020 Segmentation
Manually digitized
Training Dataset to
Train Model
Random Forest
Classification and Dissolve

Manual Inspection
and Final Landslide
Inventory
Auto Detected Landslides
Baliyo Ghar Program
 • 100 Major landslide incident collected from 2018-2021.
Landslide
Inventory • News article collected for supportive information of
landslide.
2018-2021
• GPS location verified from satellite image and contacting
to ward representative.

Baliyo Ghar Program

 3D Demonstration / Awareness Tool

Baliyo Ghar Program

 Landslide Demonstrator

Baliyo Ghar Program

 3D Demonstration / Awareness
Tool

Baliyo Ghar Program

 Landslide Susceptibility
Mapping

Baliyo Ghar Program

 POST-EARTHQUAKE LANDSLIDE
HAZARD AND RISK IN NEPAL

post-2015 EQ under Urgency and continued landslide monitoring and

observation from satellite data interpretation to study the landslide
dynamics undertaken in collaboration with IHRR/ Durham University
LANDSAT & SENTINEL

Baliyo Ghar Program

  Key Findings
EVOLUTION OF LANDSLIDE HAZARD

• Landslide hazard in the 14 worst-affected districts

remains significantly higher in 2020 than on the day of the
earthquake in 2015 (Rosser et al., 2021; Kincey et al., 2021).

• The spatial pattern and density of landsliding has evolved

through time, with systematic upslope and northward shifts
(Kincey et al., 2021).

• Coseismic landslide locations tend to

persist within mapped post-seismic
inventories, but a wider population of newly activated,
spatially dispersed landslides has also developed after the
earthquake (Kincey et al., 2021).

Baliyo Ghar Program

 Key Findings
 LANDSLIDE INVENTORIES AND MAPPING
TECHNIQUES

• At present, many forms of landslide assessment are

too slow to generate relative to the speed of a
humanitarian response, despite increasingly rapid
access to high-quality imagery (Williams et al., 2018).

• The valueof information on

landslides evolves rapidly as a disaster
response develops, so identifying the purpose,
timescales, and end users of a post-earthquake
landslide assessment is essential to inform the
approach taken (Williams et al., 2018).

• Automated approaches to
generating landslide inventories are improving but
manual mapping by trained analysts still remains the
most accurate method. However, recent advances in
cloud-based computing of time series satellite
imagery holds considerable potential for future
automated landslide mapping.

Baliyo Ghar Program

 METEOR: Objectives

I. Delivery and use of open-source national-scale exposure

datasets
II. Uptake of protocols to develop critical exposure information
from EO data;
III. Uptake of sustainable business model for exposure
application in DRM by stakeholders;
IV. Exposure data integrated into national DRM policy and
planning (where possible in a three year project);
V. Improved capacity
VI. Creation of a network of stakeholders
download: https://meteor-project.org/data/; visualisation:
https://maps.meteor-project.org
Baliyo Ghar Program
Baliyo Ghar Program
 Meteor Data In Bipad Portal
Platform

Baliyo Ghar Program

 Risk Scenarios development with IHRR/DU for
Scenario Exercise of Nepal and US Army

• Population and building

exposure in Nepal.

• Total population and

number of residential
buildings by
construction type within
each VDC in Nepal
according to the
National Population and
Housing Census (2011).

Baliyo Ghar Program

 DesInventar Data Summary (1971-2020)
Year 50
No. of Records 39,467
Death 48,333
Missing 4,060
Injured 339,236
Affected People 8,255,739
Destroyed House 858,610
Affected House 544,371
Loss Value 7,097,906,169,647

Baliyo Ghar Program

 Source

• https://nepal2015eq.webspace.durham.ac.uk/
key-findings/
• https://meteor-project.org/documents/
• https://www.pnas.org/doi/pdf/10.1073/pnas.1807
433115

Baliyo Ghar Program