GEM GLOBAL MOSAIC OF RISK MODELS

North and Sub-Saharan Africa (NSSA)

VIEW

1. Global Risk Modeling – Data and Results for Africa

Within the framework of GEM’s Global Earthquake Risk Model, the components necessary for the estimation of economic and human losses were developed for 56 countries in Africa. This information represents the most comprehensive resource for seismic hazard and risk assessment in the region. It comprises the characterization of seismic hazard, the description of the physical building stock, and definition of  vulnerability and fragility functions. The seismic hazard model for Africa was delivered in 4 separate projects covering Eastern Africa, Southern Africa, West and Central Africa and North Africa. Three occupancy classes were considered in modeling the physical building stock: residential, commercial and industrial. Population exposed to ground shaking is modeled taking into account occupancies at different times of the day (e.g. day, night and transient hours). Fragility and vulnerability functions are described according to the probability of loss ratio for all the predominant building typologies identified in Africa.

2. Exposure Modeling

For most countries in Africa, the majority of the population live in rural areas in contrast with the rest of the world where the building stock, infrastructure and population are mainly concentrated in urban areas. For example, the most recent census conducted in Ethiopia in 2007 estimated about 81% of the Ethiopians live in rural areas. The settlement of a household is usually related to their economic condition and the type of house they live in. To assess and calibrate earthquake losses for each country in Africa, we have used data from population and housing censuses (at the smallest available administrative level) which capture demographic variations in a country. They are the most complete and updated databases containing information about population and households.

2.1 Identification of variables

National Census Data describes various dimensions of demographic, economic and social data pertaining to persons within a country. To model the building stock exposed to ground shaking, we collected the variables relevant to describe the main material of construction, lateral load resisting system, number of storeys, and less frequently, also the date of construction. Type of foundation, roof type, floor type, vertical and horizontal irregularities were collected (when available) to support assigning vulnerability classes to assets.  Secondary information regarding the built-up area, construction cost and household size was also collected from the existing literature, technical reports, national housing census and the judgement of local experts.

2.2 Mapping and Estimating Buildings

A key step in developing a reliable exposure model is the mapping of building classes to attributes featured by the housing census and the socio-economic data. The three main steps involved in the mapping are:

  • Identification of the most common building classes using the GEM taxonomy.

  • Association of each building class with the attributes used by the local dataset.

  • Definition of the fraction of each building class.

 

Based on information available for each country and the procedure described above, all dwellings were classified according to the predefined building classes. In total, over 30 unique building classes were identified for Africa. The number of dwellings were converted to the number of buildings using the number of storeys per building class, average number of households per storey, and the type of area (urban, rural or large city).

2.3 Average Area and Replacement Cost

To understand and estimate the total economic losses due to direct damage, we assigned an average area and replacement cost using construction manuals and technical documentation available for the region. The average area and replacement cost were assigned according to building class, number of stories and area type (urban, rural or large city). The replacement costs were further disaggregated into costs for structural components, nonstructural components and contents.

The table below presents the basic statistics from the exposure model for each country or territory in Africa.

Country
Population (thousands)
Dwellings (thousands)
Commercial (thousands)
Industrial (thousands)
Algeria
41,078
6,768
363
111
Angola
25,790
5,545
273
68
Benin
11,133
1,211
107
31
Botswana
2,035
479
11
3
Burkina Faso
19,633
3,757
149
33
Burundi
8,054
1,686
7
2
Cameroon
19,407
3,392
139
29
Cape Verde
492
115
5
2
Central African Republic
4,660
772
45
23
Chad
14,900
2,641
121
42
Comoros
814
145
3
2
Congo
5,261
611
10
9
Democratic Republic of the Congo
50,028
5,439
356
140
Djibouti
819
87
11
3
Egypt
94,799
22,169
363
129
Equatorial Guinea
1,268
209
4
3
Eritrea
5,618
894
60
14
Eswatini
1,368
323
4
2
Ethiopia
104,958
21,420
259
88
Gabon
2,026
336
5
4
Gambia
2,828
213
5
5
Ghana
24,676
5,884
224
46
Guinea
10,524
1,472
29
25
Guinea-Bissau
1,862
249
5
4
Ivory Coast
22,717
4,180
64
53
Kenya
35,738
8,739
229
80
Lesotho
1,946
423
14
7
Liberia
3,478
671
22
4
Libya
5,173
960
61
31
Madagascar
22,435
4,774
236
64
Malawi
13,078
2,893
95
15
Mali
14,529
2,356
48
24
Mauritania
3,538
553
10
3
Mauritius
1,265
560
17
14
Mayotte
254
57
1
1
Morocco
33,611
7,265
195
62
Mozambique
21,944
4,836
270
61
Namibia
2,114
465
16
4
Niger
21,478
3,384
180
41
Nigeria
140,432
28,198
1,719
359
Reunion
877
128
3
2
Rwanda
10,379
2,425
30
9
Sao Tome and Principe
205
42
1
1
Senegal
13,927
1,608
42
51
Seychelles
96
23
2
1
Sierra Leone
6,998
1,247
40
4
Somalia
14,743
2,382
82
31
South Africa
55,678
16,939
387
110
South Sudan
8,261
1,114
88
26
Sudan
30,894
6,181
93
21
Tanzania
44,050
9,277
299
40
Togo
6,192
1,299
18
15
Tunisia
11,155
3,274
97
29
Uganda
34,647
7,349
158
25
Zambia
13,060
2,676
75
16
Zimbabwe
12,675
2,966
76
17
building dist est cap value.png

Africa Exposure Map

3. Vulnerability

The vulnerability component characterizes the likelihood to suffer damage or loss given a hazard intensity. The relation between probability of loss and hazard intensity is expressed by a vulnerability function, whilst the relation between probability of damage for each damage state and hazard intensity is represented by a fragility function. Despite the notable advances in regional seismic vulnerability modelling in the last three decades, a uniform set of vulnerability or fragility functions covering all of the building classes in Africa was not available. Moreover, with a few exceptions, most of the existing vulnerability functions have not been tested against damage data from previous events and have not been applied within a probabilistic framework for earthquake loss assessment. In general, this approach relies on the following steps:

 

  1. Identification of the most common building classes in the region, using peer-reviewed literature, web surveys (https://platform.openquake.org/building-class/), and World Housing Encyclopedia reports.
     

  2. Development of simplified numerical models for each building class, using data from the literature and results from experimental campaigns (e.g. yield and ultimate global drift, elastic and yield period of the first mode of vibration, participation factor of the first mode of vibration, common failure mechanisms). Some of the building classes had to be explicitly modelled using complex 3D models due to the lack of information in the literature.
     

  3. Selection of ground motion records using local strong motion databases, and considering the local seismicity and tectonic environment. To this end, seismic hazard disaggregation at the location of the most urbanized centers supported the identification of the combinations of magnitude and distance, which contribute the most to the seismic hazard. The use of a large set of actual time histories aims at propagating the record-to-record variability to the vulnerability assessment.
     

  4. Performing nonlinear time history analysis to evaluate the structural response (i.e. engineering demand parameter (EDP) – maximum displacement and acceleration) of the simplified numerical model against the selected ground motion records. This step uses the open-source package for structural analysis OpenSees, and the Risk Modelers Toolkit developed and supported by GEM.
     

  5. Evaluation of the structural responses of the numerical models in order to evaluate the evolution of damage with increasing hazard intensities. In this process, the probability of exceeding each damage state for a set of intensity measure levels is defined (i.e. fragility functions).
     

  6. The fragility functions can be converted into vulnerability functions (i.e. probability of loss ratio conditional on ground shaking) using a damage-to-loss model. Such functions can be used directly in the assessment of economic and human losses due to earthquakes.

 

This framework is supported by a set of tools that can be improved upon the release of new models and datasets. As an example, fragility models for the four most common building classes in Africa are illustrated below.

africa vul1.png
africa vul3.png
africa vul2.png
africa vul4.png

4. Seismic Hazard

The seismic hazard for the African continent was developed considering four regions. The SSAHARA project encompasses countries in eastern Sub-Saharan Africa. The North Africa and West Africa models were developed internally at GEM with collaborations from local experts in the respective regions. For southern Africa, the Council of Geoscience of South Africa developed and shared a seismic hazard model for the region. Below are the links to the datasets and results of all four models and associated technical documentation.

  1. Eastern Sub-Saharan Africa

  2. North Africa

  3. West Africa

  4. South Africa

Africa Hazard Map

5. Seismic Risk Results

The seismic hazard of the African continent is generally considered to be low to moderate. However, the physical vulnerability of the built environment is relatively high, which exacerbates the impact of ground shaking. In order to understand and develop pragmatic solutions towards seismic risk reduction, it is imperative to estimate accurately the possible losses. The OpenQuake-engine was used to estimate a number of risk metrics for each country. The loss parameters include annualized average loss, loss ratio and probable maximum loss for different return periods. Moreover, as part of GEM’s initiatives to increase risk awareness, a country risk profile was developed for each country in Africa (https://www.globalquakemodel.org/country-risk-profiles). Below are figures showing the top 20 countries in Africa in terms of average annual losses and their corresponding loss ratios.

africa aal new.png

6. Partners and Contributors

The Africa seismic risk model extensively relies on the enthusiasm and commitment of various organizations that openly collaborated with GEM and its partners. The creation of this model would not have been possible without the support provided by many experts. A list of the individuals that contributed to the development of the Africa seismic risk model is provided below.

Name
Institution
Country
Abdourahman Houmed-Gaba Maki
Intergovernmental Authority on Development
Republic of Djibouti
Abilio Carlos Langa
National Disasters Management Institute
Mozambique
Ahmed Essam
Understanding and Managing Extremes Graduate School Pavia
Egypt
Amina Varachia
Cape Town's Disaster Risk Management Centre
South Africa
Ansie Smit
Natural Hazard Centre, University of Pretoria
South Africa
Carlien Bou-Chedid
Independent Consultant
Ghana
Catherine Ahimbisibwe
Prime Minister Office
Uganda
Chimwemwe Bob Banda
Department of Disaster Management Affairs
Malawi
Crispin Kinabo
University of Dar El salaam
Tanzania
Elmien Steyn
Cape Town's Disaster Risk Management Centre
South Africa
Eward Boniface
Disaster Management Department of the Prime Minister's Office
Tanzania
Harouna Nshimiyimana
Rwanda Housing Authority
Rwanda
Hassan Mdala
Department of Disaster Management Affairs
Malawi
Herve Villard
University of Rwanda
Rwanda
Isa Lugayizi
Ministry of Energy & Mineral Development
Uganda
Joao Alberto Mugabe
Eduardo Mondlane University, Maputo
Mozambique
Joseph Stephen Mayunga
Ardhi University
Tanzania
Keflemariam Sebhatu
Intergovernmental Authority on Development
Republic of Djibouti
Titus A. Kuuyuor
United Nations Development Programme
Mozambique