BRIEF October 12, 2018

World Bank’s preliminary assessment of disaster affected areas in Central Sulawesi using the GRADE methodology

Background: As part of the World Bank’s response to the Sulawesi earthquake and tsunami, a rapid assessment of the damage‑affected areas in Central Sulawesi was conducted using the Global Rapid post‑disaster Damage Estimation (GRADE) methodology. The findings of the report will be shared by World Bank Group CEO, Kristalina Georgieva, during her visit to Sulawesi on October 12th, 2018. The following are relevant points related to the report.


Report and Findings

1.  Who produced this report (is this a World Bank product)?

The assessment was conducted by the World Bank’s Disaster-Resilience Analytics and Solutions (D-RAS) team. This is a team of technical experts that provides advisory and analytical services, and develops custom-built tools and solutions in the area of disaster risk management. The team received financial support through the Global Facility for Disaster Reduction and Recovery (GFDRR).  

2.  Were GoI counterparts and/or donor agencies consulted?

The GRADE approach is a remote, desk-based, rapid damage assessment method deployed on request soon after a disaster. The approach adopts evolving and innovative natural hazard risk modeling technology in order to rapidly fulfill post-event damage assessment requirements. The methodology underpinning the assessment is freely available.

Due to the remote nature of the assessment, consultations were not conducted. However, locally published datasets available online, including from the National Disaster Management Authority (BNPB), BPS (Statistics Agency Indonesia) and others were referenced and incorporated. Further validation with Government authorities and data collected in the field will be required.  

3.    Can you summarize three key findings of this report?

  • The total economic damages are estimated at over US$500 million (~IDR 8 trillion). Of this, approximately US$180 million (IDR 2.7 trillion) is expected to be in the housing sector; US$185 million (IDR 2.82 trillion) for commercial/industrial buildings; and US$165 million (IDR 2.5 trillion) for infrastructure. Losses to all equipment external to these assets (e.g. cars) is not covered by the assessment.
  • Key affected sectors include housing, commercial and / or industrial buildings, and public and private infrastructure.
  • The high impact on commercial / industrial buildings could affect operations and recovery in the retail and tourism, education and health sectors.  Government public buildings were also affected. The impact on school-university buildings was also considerable.

4.       What is the main benefit of this type of report?

The main benefit is the speed at which the damage estimation can be produced. Within 10-14 days of an event, stakeholders can have access to loss estimates and spatial distribution of damages. This can support the development of post-disaster recovery and reconstruction strategies, and inform appropriate, timely, and efficient courses of action.

For the Sulawesi earthquake and tsunami, as of October 9, 2018, this is the first report to produce sector-based preliminary economic loss estimates, based on scientific, economic and engineering data and analysis, which can inform disaster recovery and reconstruction processes.


Assessment and Methodology

5.       What data was used in this assessment?

GRADE is based on an open loss modeling approach developed by the World Bank’s Disaster-Resilience Analytics and Solutions (D-RAS) team, that includes:

  • Analysis of satellite imagery and other ground‑collected data (BNPB, Ministry of Education, PUPR, AHA Centre, media etc.)
  • Remote-sensing imagery (UNOSAT, COPERNICUS, DigitalGlobe, Google, HOTOSM, MapAction),
  • Information coming out of early assessments, as well as social media data for results calibration.
  • Spatial characteristics developed for tsunami events which included inundation extent and ground deformation analysis that were produced based on a remote sensing damage assessment.

6.       What kind of assessment is this, how was it conducted, and what is the main benefit?

The GRADE approach is a remote, desk-based, rapid damage assessment method deployed on request soon after a disaster. The approach adopts evolving and innovative natural hazard risk modeling technology in order to rapidly fulfill post-event damage assessment requirements. The approach prioritizes the housing and infrastructure sectors, followed by other sectors, such as agricultural production, as needed. The GRADE approach and outputs are intended to create an independent, credible sectoral quantification of the spatial extent and severity of a disaster’s physical impact.

7.       How certain/uncertain are the figures presented?

The assessment carries a significant degree of reliability. The GRADE method was used following seven disasters that occurred between April 2015 and June 2018, with “like for like” field estimations accuracy when compared with subsequent and more-detailed post-disaster analyses. The results are calibrated against an inflow of consequence data (remote sensing, drone footage, social media videos, early assessments, crowdsourced info). However, GRADE’s outputs are still estimates arising from remote-based calculations that are influenced and updated from available ground-based data. While there is confidence in the overall economic estimates and distribution of damage, the confidence level at the individual asset level is low as these results are based on projections from the latest census data that may be outdated. Overall, the results are influenced by availability, accuracy, vintage, socioeconomic/political sensitivities of baseline exposure, and the flow of damage data during the early post-disaster period.

8.       How can this complement government or other assessments?

The GRADE assessment can inform and serve as inputs to the detailed and exhaustive Government-led damage and loss assessment processes. The GRADE assessment can provide decision-makers with a first order of the economic impact in order to gauge the magnitude of the event’s consequences, identify priority sectors for reconstruction, provide information on differential geographic impacts, and inform on relative public versus private sector damages. Furthermore, the assessment can contribute to sectoral baseline information needed for the design of rehabilitation and reconstruction-recovery plans. In this way, the results could support the design of a short-term plan to re-establish affected services and to stabilize conditions of affected populations through temporary measures. It can also provide information for investment plans and for intervention strategies for the recovery and reconstruction of damaged infrastructure.

9.       What are some limitations of this assessment?

This assessment does not account for the social component. The impacts of business interruption and land value are also not captured. The assessment looks mainly at the direct physical damage to buildings and infrastructure, which is crucial information needed in order to quantify and plan reconstruction efforts. In this case, in Palu, it also covers the capital stock (e.g. agricultural stock and irrigation channeling) but does not cover the land loss of ca. 700ha.

10.   Has this assessment been used for other countries; is it standard practice of the World Bank to conduct these kinds of assessments? Can you give examples of how other countries used this kind of assessment?

Every disaster is unique, and the World Bank does not conduct a rapid assessment for each disaster; rather it focuses on cases where it can provide added value to Government counterparts.

The GRADE approach has been successfully used for seven recent disasters, including Madagascar (after Cyclone Enawo in March 2017 and Cyclone Awa in January 2018), Haiti (after Hurricane Matthew in October 2016), Ecuador (after the earthquake on April 16, 2016), Nepal (after the earthquake on April 25, 2015), Dominica (after cyclone Maria, September 2017) and Fuego volcanic eruption in Guatemala (on June 3, 2018).  For example, after Cyclone Enawo in Madagascar (March 2017), the outputs from GRADE provided swift assessments that informed the preparation and implementation of disaster relief and emergency response strategies, improving the effectiveness of the response. Also, after Hurricane Matthew in Haiti (October 2016), the outputs helped develop the rapid impact estimates, which, in turn, was used by the International Monetary Fund (IMF) to determine whether it should trigger its post-crisis mechanism for the country.