Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

GIS based quantification and mapping of climate change vulnerability hotspots in Addis Ababa

GIS based quantification and mapping of climate change vulnerability hotspots in Addis Ababa Background: Unlike the studies undertaken on agricultural and hydrological sectors, focused climate change vulnerability researches in urban centers in Ethiopia is not widely available and of recent history. However, as many signals of climate change vulnerability started to happen in urban centers as well, it is inevitable to analyze, quantify, map, prioritize and be prepared for adaptation measures. This study is therefore, tried to assess, quantify and map climate change vulnerability in Addis Ababa, Ethiopia, by integrating two climate change vulnerability assessment models, namely, the Sullivan and Meigh’s Model of composite climate change vulnerability index and the IPCC’s approach of vulnerability assessment which comprises exposure, sensitivity and adaptive capacity. Fifteen sub- components of vulnerability indicators were identified in ten sub-cities (Addis Ketema, Arada, Akaki-Kality, Bole, Gulelle, Kirkos, Kolfe-Keraniyo, Lideta, Nifasilk-Lafto and Yeka) of Addis Ababa. Due to the scale, degree, amount and unit of measurement for the selected indicators varied, their values were normalized to a number which ranges between 0 and 1, indicating as the values increased to 1, vulnerability to climate change increases. The study uses Iyengar and Sudarshan’s unequal weighting system, to assign a weight to all indicators. The results were mapped using ArcGIS 10.2 package. Results: The results indicated that the ten sub-cities in Addis Ababa were found in different levels of vulnerability to climate change. The exposure and sensitivity were highest for Addis Ketema, Arada, and Lideta which are found in central parts of the city, with a normalization index value greater than 0.5. The adaptive capacity index is the highest in Gulelle, Bole, and Arada sub-cities. These sub-cities have better quality houses, well-planned districts, good infrastructural facilities and good coverage of green areas compared to others. The overall climate change vulnerability was the highest (normalized index > 0.5) in Arada, Addis Ketema and Lideta, due to the adaptation capacity is the lowest compared to other sub-cities. Conclusion: Addis Ababa is vulnerable to climate change impacts and the degree of vulnerability is underpinned by the interaction of multiple factors mainly adaptive capacities of sub-cities, location based characteristics and changes in climatic parameters. These present a need to strengthen mitigation and adaptation activities and prioritize sub cities for intervention based on the degree of vulnerability. It is also understood that the Sullivan and Meigh’sModel andIPCC’s approach for climate change analyses, could be used simultaneously for preparing vulnerability index per different geographical locations. Keywords: Climate change, Exposure, Sensitivity, Sullivanand Meighmodel,Vulnerability index * Correspondence: getyonani@yahoo.com Ethiopian Institute of Architecture, Building Construction and City Development (EiABC), Environmental Planning Programme, Addis Ababa University, P.O. Box 518, Dej. Balcha Aba Nefso St, Addis Ababa, Ethiopia Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 2 of 17 Background resilience reduces vulnerability (Folke et al. 2002). Vulner- Climate change is a major problem in the world in the ability to a climatic stressor is essentially a composite of ex- current century and it is the biggest and greatest challenge posure, a degree of sensitivity to the stressor, and the ability that humanity faced in the twenty-first Century. The recent of the exposed system to cope with the stressor. Vulner- fifth Assessment Report of the Intergovernmental Panel on ability is mostly considered to be a local phenomenon that Climate Change (IPCC 2014) indicated that each of the last is investigated by local-scale studies in which the specific three decades has been successively warmer at the Earth’s societal (and physical) situation can be taken into account surface than any preceding decade since 1850. The globally much better than in a global-scale study(Doll 2009). averaged combined land and ocean surface temperature Mapping vulnerability hotspots of climate change in data as calculated by a linear trend show a warming of 0.85 urban areas underpin climate change adaptation and miti- [0.65 to 1.06] °C over the period 1880 to 2012. It is well ac- gation policies and strategies in the contemporary world. It cepted that climate change will have a far more detrimental helps to figure out which places and people are the most effect on developing countries compared to developed vulnerable, as well as the degree of vulnerability and pos- countries; this is mainly because the capacity to respond to sible adaptation options (Gebreegziabher et al. 2016). Due such changes is the lowest in developing countries. More- to the potential of maps in depicting vulnerability hot spots over, it seems clear that vulnerability to climate change is is high, as it has strong visual elements and can easily be closely related to poverty, as the poor are least able to interpreted than text, its usage is increasing in different respond to climatic stimuli (Srivastava 2015). The case is research groups and organizations (de Sherbinin 2014). also true for Africa, as the continent is viewed as the most More specifically, vulnerability mapping helps to set vulnerable region to climate variability and change. Region- three policy measures. First, it is used to specify long-term ally in East Africa studies indicate that in countries like targets for the mitigation of climate change; second, to Burundi, Kenya, Sudan and Tanzania people are hit high identify vulnerable places and people and to prioritize re- by the impacts of climate change (Hassaan et al. 2017; source allocation for adaptation and to put adaptation Mwangi and Mutua 2015; Shemsanga et al. 2010). In measures (Fussel and Klein 2006). In climate change Ethiopia, within the last decades, high temperature values literature, two components of vulnerability are identified: recorded in different parts of the country. Various General internal and external dimensions (Gebreegziabher et al. Circulation Models (GCMs) results also tell as the future 2016). The internal dimension refers to frailty, insecurity, change climatic elements will be high. For instance, the and capacity to anticipate, cope with, resist and recover average future change for the whole of Ethiopia for three from adverse effects of shocks. The external dimension 30-year periods with A2 emissions shows warming in all involves exposure to risks and shocks. four seasons in all regions, with annual warming in In Ethiopia, the impact of climate change is studied Ethiopia by the 2020s of 1.2 °C with a range of 0.7–2.3 °C, largely at country level in different sectors mainly em- 2050s by 2.2 °C, range 1.4–2.9 °C (Conway and Schipper phasizing on the impacts of agricultural productivity and 2011). So far some studies predict long-term future climate water sectors. These studies reveal that climate change change situations that could prevail up to the end of this and variability severely affect economic conditions liveli- centuryinEthiopia(McSweeney etal. 2010). hoods and agricultural production and food security, If the occurrence of climate change is real, it is a must to water resources and health (Gebregiziabhre et al. 2016, identify it’s possible impacts and prepared for better adap- Legese et al. 2016, Tafesse et al. 2016, Melese 2016, tation mechanisms. In view of this, it is important to Abebe and Kebede 2017, Amare and Simane 2017). The understand the concept of climate change vulnerability, impact of climate change in urban scale is also studied identify indicators, quantify and map in order to have a in towns like Dire Dawa (Billi et al. 2015), Addis Ababa better understanding and to implement quick interven- (CLUVA 2011; Cochrane and Costolanski 2013; Birhanu tions. Climate change is not a single concept by itself and et al. 2016; Feyissa et al. 2018) indicating the identified rather it enclosed various concepts within it. In most cases, cities are vulnerable to climate change impacts mainly vulnerability, resilience, and adaptive capacity are the flooding and urban heat. central concepts for the analyses and understanding of the Though these studies provide some useful insights in impacts of climate change, because, together they provide a the area of vulnerability to climate change, and reflect framework that links biophysical climate sensitivity to the current efforts to understand the relationship social and economic factors that mitigate or amplify the between climate change and vulnerability, their scale of consequences of environmental changes. Vulnerability as- study is either at national level which is very coarse or at sessment and resilience is mainly analyzed with the topics case-specific level which do not indicate the overall like sustainability, hazards and climate change impacts vulnerability conditions. This study is, therefore analyzes (Malone 2009). Low resilience systems are intrinsically vul- how compiled bio-physical and socio-economic as well nerable to stress and shock, so in this sense increasing as geographic factors determine the vulnerability to Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 3 of 17 climate change in Addis Ababa, under different sub ad- its strategic role within the overall economic develop- ministration level (sub-cities). It analyses and construct ment of the country (World Bank 2015). vulnerabilty index at sub-city level, identify and map vul- The city is located in the central highlands of Ethiopia nerability hotspots. It also tries to know whether the vul- covering an areal extent of about 527 km with an aver- nerability to climate change is spatially varied or not in age elevation of 2600 m above mean sea level (asl). The one large city. In this regard, there is no prior studies altitude range extends from the highest peak at Mount available which integrates the IPCC’s approach and Entoto which is 3041 m to 2051 mean above sea level at Sullivan and Meigh’s model for vulnerability assessmnet the lower part of Akaki plain. The topography is undu- in Addis Ababa. The result of the study helps as an in- lating and form plateau in the northern, western and put for planners and different organizations which southwestern parts of the city. Bole, Akaki, and south already starts imposing a number of policies, strategies western part of the city is characterized by gentle and programs aimed at enhancing the adaptive capacity morphology and flat land areas. An average maximum and reducing the vulnerability of the country to climate temperature for Addis Ababa in the last 60 years is variability and change. 22.9 °C and the average minimum temperature is 10.2 °C. The average annual rainfall for Addis Ababa is 1184 mm. The wet season is from June to mid-September. The Methods urban area is endowed with three major rivers: Kebena, Study area description Little Akaki and Big Akaki rivers and numerous small Addis Ababa is found between 8°50′ N to 9°5′ N and streams. The population density varies from sub-city to 38°38′ E to 38°54′ E. It is the capital and the largest city sub-city. The highest density in Addis Ketema sub city of Ethiopia, with a total projected population of 3.44 (37,215p/sk.km) to the sparse density in Akaki-Kality million people in 2017(Federal Democratic Republic of sub-city (1832p/sk.km). All the sub-cities in the down- Ethiopia Central Statistical Agency 2013) and the admin- town have high population density compared to sub cit- istration of the city is divided into ten sub-cities. Addis ies found in peripheral areas. Fast rate of urban Ababa is home to 25% of the urban population in expansion and built up area are rapidly increasing in Ethiopia and is one of the fastest growing cities in Addis Ababa (Woldegerima et al. 2016), which were Africa. The city is a major growth corridor for the coun- largely attributed to population pressure on the land, a tries vision to become a middle income by 2025. The rapidly growing infrastructure and poor land use plan- city alone currently contributes approximately 50% to- ning (Teferi and Abraha 2017).The study area is pre- wards the national Gross Domestic Product, highlighting sented in Fig. 1. Fig. 1 Sub city boundary and Elevation map of Addis Ababa Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 4 of 17 Climate change vulnerability assessment approach climate change vulnerability. This study uses indicator ap- Various vulnerability assessment methods are available proach by integrating Climate change Vulnerability Index for urban areas. For instance, the Climate change and (CVI) model developed by Sullivan to construct climate Urban Vulnerability in Africa (CLUVA) project uses the change vulnerability index. This index is used as a checkup vulnerability assessment for Addis Ababa based on the in indicator approach, whereas the weight is assigned in works done by Moser et al. 2010, based on asset, institu- Arc GIS analyses. tion, attitude and physical as a collective umbrella of Vulnerability assessment follows the standardized vulnerability. work done by Sullivan and adopted by many types of According to the IPCC’s definition, vulnerability to research due to it comprises various physical and socio- climate change and variability is represented by three economic data for urban areas. The methodology used elements: exposure, sensitivity, and adaptive capacity. for CVI is based on the methodology of Water Poverty Exposure can be interpreted as the direct danger (i.e., Index developed by Sullivan and Meigh (2005) as speci- the stressor), and the nature and extent of changes to a fied in eq. 3. The main components and subcomponents region’s climate variables (e.g., temperature, precipita- of vulnerability index developed by Sullivan and Meigh tion, extreme weather events). Sensitivity describes the is presented in Table 1. human–environmental conditions that can worsen the w R þ w A þ w C þ w U þ w E þ w G hazard, ameliorate the hazard, or trigger an impact. r a c u e g CVI ¼ ð3Þ Adaptive capacity represents the potential to implement w þ w þ w þ w þ w þ w r a c u e g adaptation measures that help avert potential impacts. Adaptive capacity is considered “a function of wealth, R(Resources), A(Access), U(Use), C(Capacity), E(Envir- technology, education, information, skills, and infra- onment), G(Geospatial), and w,w ,w ,w ,w ,w – the r a u c e g structure, access to resources, and stability and manage- weights of indicators. ment capabilities” (Adger et al. 2005). However, slight modification can be done on the Therefore, analyzing vulnerability must involve identi- Sullivan model based on the actual context of the study fying not only the threat but also the “resilience,” or the area. Considering the basic indicators of climate change potential responsiveness of the system and its ability to exploit opportunities and resist or recover from the Table 1 Major Components of CVI (Based on Sullivan and negative effects of a changing environment. The first Meigh 2005) two components together represent the potential impact CVI Component Sub Indicators and adaptive capacity is the extent to which these im- Resource Total water resource pacts can be averted. Thus vulnerability is a potential Access Clean water impact (I) minus adaptive capacity (AC). Sanitation This leads to the following mathematical equation for Capacity Education level of a population vulnerability (International Crops Research Institute for Under five mortality rate the Semi-Arid Tropics 2009). Income V ¼ I‐AC ð1Þ People live in informal housing Which can be elaborated as follows (Moss et al. 2001): Access to a place of safety in the event of flooding Existence of disaster warning system Vulnerability ¼ðÞ adaptive capacity ‐ðÞ sensitivity þ exposure Use Domestic water use ð2Þ Industrial water use This study uses eqs. 1 and 2 which stated above, and Agricultural water use differentiates the indicators into three categories as Environment Water stress exposure, sensitivity and adaptive capacity. Water management Population density Climate change vulnerability index Loss of habitat Quantitative assessment of impact (vulnerability) is usually Geospatial Deforestation done by constructing a vulnerability index. This index is based on several sets of indicators that result in the vulner- Flood ability of a region. It produces a single number, which can Infrastructure be used to compare different regions (International Crops Thermal Heat Index Research Institute for the Semi-Arid Tropics 2009). Vari- Land conversion from natural vegetation ous indexes have been developed to construct and map Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 5 of 17 vulnerability with minor modifications in Sullivan and with vulnerability, the above formula will be changed to Meigh Model 2005, the methodology followed is integrat- the following as described in eq. 5. ing IPCC’s definition of vulnerability, in three sub layers Max X ‐X ij ij of exposure, sensitivity and adaptive capacity. Every com- X ¼ ð5Þ ij Max X ‐ Min X ponent is made up of sub-components. Based on these ij ij components, the 15 selected indicators presented on Once the normalized scores have been completed for Table 2 were identified and analyzed for Addis Ababa. entire selected indicators the simple average construct vulnerability index by adding all normalized scores (eq. Normalization of the indexes 6). Normalization of the indicators will undertake to ensure the P P comparability of the indicators. This was carried out using x þ y ij j j ij the methodology developed for the calculation of the Hu- VI ¼ ð6Þ man Development Index (UNDP 2006)using eq. 4.Then all the indicators assigned with values range from 0 to 1. Where VI is vulnerability index, i represent the sub-city and j represents indicator and K is the total number of indicators. X ‐ Min X ij ij X ¼ ð4Þ The value of the districts (i) is sub-cities and j (indicators) ij Max X ‐ Min X ij ij is replaced with normalized CVI results in this study. Where X is the separated value in the distribution, Assigning weights to indicators Min (X ) is the minimum value in the distribution and The basic challenge in constructing indices is the lack of ij Max (X ) is the maximum value of the mean of the dis- standard ways of assigning weight to each indicator. The ij tribution i is the sub-city and j is number of indicators. two most common weighting methods used to combine The value of the normalized equation falls between 0 indicators are equal and unequal weighting schemes (Geb- and 1. Where, 1 being the highest value and 0 with being reegziabher et al. 2016). The present study uses an unequal the least vulnerable area for the indicators with positive method of Iyengar and Sudarshan’s 1982 to give weight to relationship with climate changes vulnerability. Unless if all indicators. The choice of the weights in this manner the indicators are assumed to be negative relationship would ensure that large variation in any one of the Table 2 Selected Indicators of Exposure, Sensitivity and Adaptive capacity and their data source Component Indicators Functional Relation-ship Layers in Sullivan Definition Data Source Exposure Layers Mean air temperature change + Geospatial National Meteorological Agency Change in Land surface Temperature + Geospatial Extracted from Landsat Flood Risk + Geospatial Constructed from different layers Sensitivity Layers Mud(wood) house type + Capacity Central Statistical Agency Population density + Environment Central Statistical Agency Vegetation cover – Environment Extracted from 2017 Landsat Adaptive capacity Layers Unemployment rate + Capacity Central Statistical Agency Literacy rate – Capacity Central Statistical Agency Under five mortality rate + Capacity Central Statistical Agency Activity Rate Capacity Central Statistical Agency Distance from emergency centers + Capacity Addis Ababa City Fire & Emergency Prevention & Rescue Authority Road Length + Access Addis Ababa Road Authority Access to tap water + Access Central Statistical Agency Access to toilets + Access Central Statistical Agency Distance from all type of health – Access Integrated Land Information centers and Technology Office Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 6 of 17 indicators would not unduly dominate the contribution of Under temperature change layers two sub-layers have the rest of the indicators and distort the overall rankin- been prepared. One is air temperature change layer, g(Iyengar and Sudarshan 1982). which obtained from direct measurement of meteoro- In Iyengar and Sudarshan’s method, the weights are logical stations and the other is the Land surface assumed to vary inversely as the variance over the temperature change layer, which were derived from con- regions in the respective indicators of vulnerability. secutive Landsat satellite images. Regarding the air That is, the weight w is determined by eq. 7. temperature, three stations have been selected of five stations found in Addis Ababa based on the data avail- qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi W ¼ : ð7Þ ability and geographical variability. These were Entoto, Var X i ij Addis Ababa, and Bole observatories. The record at Entoto starts from 1989 onwards. Therefore the mean Where C is a normalizing constant: (Eq. (8)) temperature record from 1989 to 2016 was divided into 2 3 two parts for all the three stations. These are from −1 j¼k (1989–2002) and from (2004–2016). The mean values of qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 4 5 ½ C ¼ ð8Þ each station have been calculated and the first 14 years Var x j¼1 i ij average was subtracted from the first mean value. Based on this, the temperature values had obtained, mapped The overall sub city index, Yi,alsovariesfrom zero (0) and reclassified. The last final result was normalized. to one (1), with 1 indicating maximum vulnerability and 0 According to this, changes in mean temperature were indicating no vulnerability at all. The higher the normal- 0.47 °C, 0.24 °C and 0.20 °C at Entoto, Addis Ababa ob- ized sub-city index, the more the level of vulnerability. servatory, and Bole respectively. The normalized scores The composite indicator for climate change vulnerability were reclassified into five groups equally distributed with factors (exposure, sensitivity and resilience) for the ith sub 0.2 normalized values range as presented in Fig. 2. Based city was obtained as: on the Iyengar and Sudarshan’s method of eq. 7, the attached weights are (0.3), (0.3) and (0.4) respectively for W Y ð9Þ i ij Land Surface Temperature Change, air temperature Where: Yi is the composite indicator of ith sub-city; change, and flood risk. The assumption was that the W is the weight for each indicator lies between 0 and 1; places with the highest temperature change are more ∑W 1 and Y is the normalized scores of indicators. vulnerable to climate change. Identified exposure layers j= ij are presented in Table 3. Zonal statistics The mean difference of LST ranges from 3.4 °C at The Zonal Statistics tool of ArcGIS was widely used in Addis Ketema to 0.8 °C at Yeka sub-city. The difference this study to assign values obtained from raster images between the 1986 and 2017 has been normalized. Both to the sub-cities. A mean value of each indicator was results from air and land surface temperature change in- computed for every zone (sub city). The average of the dicates that high rate of change is found in the northern values in each zone is assigned to all output cells in the and northwestern parts of Addis Ababa while the low same zone. Zonal statistics tool is used in classifying and elevated parts of Addis Ababa, which are sparsely popu- relating land surface in land use change in various lated have low rate and magnitude of change in studies (Youneszadeh et al. 2015; Rahmana 2016; temperature. Sierra-Soler et al. 2015). Based on the calculations, the land surface temperature changes with time and the B). Flood Risk Layers. elevation, their statistics and spatial correlations has been utilized. The application of GIS to analyze climate The flood risk is not directly obtained as a single change vulnerability has grown exponentially in the last layer as explained in temperature case. Instead of con- decade (Woodruff et al. 2017). sidering a precipitation as a direct exposure layer in an urban area, it is better to integrate it with the flood risk Result layers. In order to get flood risk layers, four sub-layers Analyses of exposure components has been weighted using multi-criteria evaluation. Based on the climate vulnerability assessment procedure These four sub-layers are a) 2017 land cover layers ob- two major exposure layers were identified. These were tained from land cover layers. The land cover layer has temperature changes and flood risk layers. These layers five land cover classes, namely built-up area; bare are widely used in climate change exposure analyses. ground, open land, vegetation cover and agricultural land. b) Slope layers: the slope is divided into five clas- A). Temperature Layers. ses 0–2%, 2–8%, 8–15%, 15–30% and greater than 31%. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 7 of 17 Fig. 2 Normalized LST and air temperature change The slope layer is classified based on FAO, the highest which is found in the Southern and North East part of the slope value with fewer floods and the least with high city. Eutric Nitisol (111.55 km ) is the second most probability to be hit by flooding. c) Drainage density dominant soil found in the central and North West part layers: The drainage data is extracted from 30 m Digital of the region. Vertisol is characterized by fine textured Elevation Model (DEM of Addis Ababa. Up to six soil with > 60% clay in composition. As a result, the stream orders have been considered for the analyses. porosity of such soil is very fine making the movement 2 2 The drainage classes are 0–0.5 km/km ,0.5–1km/km , of material difficult within the soil. Hence, the perme- 2 2 2 1–1.5km ,1.5–2km and > 2 km/km . Drainage density ability of Vertisol is very low except within the cracks has a positive relationship with flooding. that are formed during dry seasons. d) The soil layers: As surveyed by Ministry of Water and Leptosol is characterized by shallow depth underlined Energy in 2004, the soil classes in Addis Ababa are Classic by hard rock and with less developed soil. The textural Xerosols, Chromic Luvisols, Eutric Nitisols, Leptosols, class is moderately coarse-textured soil with high perme- Orthic Solonchaks and Pellic Vertisols (Gizachew Kabite: ability (Gizachew Kabite: GIS and remote sensing based GIS and remote sensing based solid waste landfill site se- solid waste landfill site selection: A case of Addis Ababa lection: A case of Addis Ababa city, Unpublished). The city, Unpublished). Soils with high permeability are dominant soil of the region is Pellic Vertisol (277.23 km ) assigned lesser weights in flood risk mapping. Based on Table 3 Exposure Layers LST (°C) 1986 LST (°C) 2017 Mean Air Temperature (°C) a b Mean Mean Difference Nor. Mean Temp (°C) Mean Temp (°C) Mean Difference Nor. Arada 24.5 27.7 3.2 0.92 16.64 16.91 0.27 0.22 Addis Ketema 24.2 27.6 3.4 1.00 16.18 16.48 0.30 0.39 Lideta 25.2 27.9 2.7 0.73 17.05 17.30 0.25 0.11 Kirkos 25.3 27.7 2.4 0.62 17.06 17.30 0.24 0.06 Nifas Silk 26.6 28.0 1.4 0.23 16.65 16.90 0.25 0.11 Akaki-Kality 27.9 29.2 1.3 0.19 16.67 16.91 0.24 0.06 Bole 26.7 28.4 1.7 0.35 16.82 17.05 0.23 0.00 Yeka 25.1 25.9 0.8 0.00 16.09 16.38 0.29 0.32 Gulelle 22.4 24.0 1.6 0.31 14.04 14.45 0.41 1.00 Kolfe 25.3 26.7 1.4 0.23 15.92 16.23 0.31 0.44 Averaged mean air temperature record from (1989–2002) interpolated from Entoto, Addis Ababa, and Bole observatories, and derived for each sub city using Zonal Statistics Averaged mean air temperature record from (2003–2016) interpolated from Entoto, Addis Ababa, and Bole observatories, and derived for each sub city using Zonal Statistics (Source: Based on data from Landsat and National Meteorological Service, constructed by authors) Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 8 of 17 Fig. 3 Layers weighted for flood risk these the highest value is assigned for Vertisol and the flood risk.The food risk map is presented in a system- least value is assigned for Leptisol. In general 30%, 25%, atic manner from high risk to very low risk as depicted 30%, and 15% weight factor is assigned for land cover, in Fig. 4. The same approach is used for flood risk map- slope, drainage density, and soil type respectively to get ping by various authors(Olatona 2017;Ouma and flood risk map of Addis Ababa. Flood layers are Tateishi 2014). presented in Fig. 3. General exposure index indicates that the highest The final flood risk map was divided into five layers, exposure were in Addis Ketema(0.72), Arada(0.66) and from low to very high flood risk categories. The area Lideta(0.5). Moderate exposure were in Kirkos(0.45), and coverage percentage of very high risk and the high-risk Gulelle(0.49).Others sub-cities have low exposure index were summed together to categorize the sub-cities. value. The least exposure was in Bole(0.135). Based on the criteria, the sub-cities fall under risk areas The flood hazard map prepared from the composite were Arada, Lideta, Addis Ketma, Kirkos and Akaki, layers of drainage density, soil, population density and which have large parts of their land fall under flood risk. slope at the exposure layer of this analysis was well inte- The final flood risk map showed that areas with highest grated with the actual data recorded on flood history. population density and lowest elevation as well as high The flood history has been collected from the Addis drainage density have a highly exposed to climate Ababa fire and emergency preparedness office from change. The upper parts of the city which is covered by 2010 to 2016. The GPS points where flood history is forest had low flood risk while the places along the recorded in the past 8 years has been taken and overlaid drainage basins and high-density areas had highest as indicated in Fig. 4. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 9 of 17 Fig. 4 Flood risk map overlaid by frequency of flood Hazard records in Addis Ababa (Source:Data from Addis Ababa Fire and Emergency Service, compiled and constructed by authors) Sensitivity components sensitive, moderately sensitive and less sensitive areas have Three layers have been identified for sensitivity compo- been identified. The sensitivity is low in sparsely populated nent based on the criteria set by different authors and areas, in open spaces and for buildings constructed by a literature. These were the population density, the house standard material. The population density, the area of type, and the vegetation cover. Similar indicators have places covered by forest and the house type share per sub been used for sensitivity analyses (Gebreegziabher et al. city is presented in Table 4. 2016; Zurovec et al. 2017). High population density was The attached weights for each the three sensitivity layers at Addis Ketema(37,215/km ) and sparsely populated at were 0.32, 0.32 and 0.35 respectively for population density, Akaki-Kality(1832) in 2017 as projected by Central Stat- forest coverage and houses constructed by mud and wood istical Authority. Regarding the house types, the housing (Fig. 5). and population census have put different house types which Based on the weights, the highest sensitivity was in include corrugated iron sheets, concrete/cement, thatch, Addis ketema, Arada and Lideta,with very high sensi- wood and mud, bamboo, plastic, asbestos, and others. For tivity index greater than 0.8, high sensitivity index is in the analyses, only house types constructed of mud and Kirkos(0.69), the medium sensitivity is in Nefassilk wood have been selected. Most parts of Addis Ababa had Lafto, Kolfe, Akaki-Kality, and Gulelle with sensitivity the houses constructed from mud and wood. Arada, Addis index from 0.4 to 0.6. Yeka and Bole had a lowest sensi- Ketema, Kirkos are the sub-cities with the majority of their tivity index(0.2–0.4). houses are constructed by mud and wood. A well-planned sub city like in Bole sub-city, the main house constriction Adaptive capacity components type is concrete. It is assumed that houses constructed by Nine adaptive capacity layers have been identified. These mud are highly sensitive to climate change and can’tresist adaptive capacity layers were categorized into three change-induced impacts like heavy rainfall than concrete sub-layers. These are socio-economic and demographic building houses. The third subcomponent of sensitivity layers, access layers, and density and distance layers. a) layer is forest cover layer. It is extracted from 2017 Landsat The socio-economic layer consists of the unemployment image. The land cover analyses indicates that Yeka and rate, activity rate, literacy rate and under-five mortality Gulelle sub cities had a high area of forest cover while rates. The data were used based on the 2007 Ethiopian Addis Ketema, Arada, Kirkos, and Lideta have a small housing and population census’s definition. The proportion of vegetation covers. Places covered by forests percentage of socio-economic and demographic layers, are less sensitive to climate change. Based on these three per sub-cities, is presented in Table 5. layers one final layer is obtained. Accordingly highly b) Access to Social Services. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 10 of 17 Table 4 Sensitivity Layers Sub City Population Density (2017 projected) Forest Area(sq.km) Houses Constructed by Wood and Mud Pop density Normalized (Km ) Normalized (−ve relation) Percent Normalized Arada 28,206 0.75 0.08 1.00 81.2 0.77 Addis Ketema 37,215 1.00 0.07 1.00 88.4 1.00 Lideta 27,483 0.72 0.02 1.00 84.3 0.87 Kirkos 18,996 0.49 0.04 1.00 76.5 0.62 Nifas Silk 6812 0.14 0.22 0.99 66.5 0.31 Akaki-Kality 1832 0.00 0.42 0.99 80.7 0.76 Bole 3283 0.04 0.41 0.99 56.7 0.00 Yeka 5292 0.10 27.35 0.00 82.7 0.82 Gulelle 10,751 0.25 15.63 0.43 87.5 0.97 Kolfe 8479 0.19 4.43 0.84 76.3 0.62 Fig. 5 Sensitivity Layers Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 11 of 17 Table 5 Percentage and normalized values of socio economic and demographic data Sub City Literacy Under five Mortality Unemployment Activity Rate (%) Normalized Rate (%) Normalized Rate (%) Normalized Rate (%) Normalized Arada 89.3 0.00 0.05 0.33 24.2 0.68 63.8 0.51 Addis Ketema 84.0 0.77 0.03 0.11 27.2 1.00 62.8 0.35 Lideta 86.0 0.48 0.04 0.16 26.2 0.89 63.4 0.45 Kirkos 88.8 0.07 0.03 0.06 21.2 0.37 64.1 0.55 Nifas silk 85.4 0.57 0.02 0.01 21 0.35 63.1 0.40 Akaki-Kality 82.4 1.00 0.05 0.31 20.3 0.27 60.5 0.00 Bole 84.1 0.75 0.02 0 17.7 0.00 67 1.00 Yeka 86.0 0.48 0.10 1 23.4 0.60 62.6 0.32 Gulelle 85.0 0.62 0.05 0.42 20.8 0.33 62.6 0.32 Kolfe 83.7 0.81 0.04 0.27 23.5 0.61 63.3 0.43 There are many access data that has to be considered for vulnerable sub cities based on the 15 climate change climate change vulnerability. However, in order to avoid vulnerability indicators(Table 8). bias, only two data, which are fully available and complete for all sub cities were considered to analyze access. These Discussions are a percentage of access to water (only access to water Climate change impact is not measured only by the ex- provided by tap both in their house or compound whether posure and sensitivity’s strength; rather, it is a matter of it is shared or private) and percentage of access to toilets adaptive capacity. The exposure layers selected for vulner- for sanitation layer. c) Distance and Density Data: The two ability analyses in Addis Ababa was lower, but the sensitiv- layers of distance (distance from the health center and dis- ity was high and adaptive capacity activities were low. tance from emergency controlling center) have been These altogether makes Addis Ababa to be vulnerable city analyzed. The health centers include both private and gov- to climate change. The distribution of exposure layers ernment health centers, all types of clinics and all types of which contains physical factors was different. For instance hospitals. The distance from emergency controlling centers the air and LST temperature changes to the central and data indicates that the average distance of each place in the northern part of the city is high, indicating the higher of sub-city from the Addis Ababa fire and emergency control- the sensitivity and the lower value of adaptive capacity in ling center. The road density layer data was obtained by that parts. Addis Ketema and Arada sub cities, which con- adding all types of road lengths and dividing it to the area tained the oldest buildings within them, have old roofed of the sub-city. The types of roads considered in these houses, limited green area coverage’s and poorly managed analyses are asphalt, large stone, gravel, earth, and cobble streets with no street trees. These all makes the exposure stone. Thedistanceand densitylayersare presentedin to climate change in this part to be high. Fig. 6. Details of the adaptive capacity data with their In contrast to this, the sub-cities with high rise building, normalized value per sub-cities are given in Table 6. but with more planned parcels of land like in Bole and Finally as depicted on Table 7, the weights has been most parts of Yeka had lower-temperature change. Another assigned to the indicators. important and influential exposure indicator which has The final adaptive capacity index indicates Bole(0.56) high factor in Addis Ababa was flooding. Floods, the most was the highest sub city in adaptive capacity. Kolfe(0.48), prevalent of natural risks, are anticipated to happen more Addis Ketema(0.43), Kirkos(0.43) and Nifas Silk(0.41) strictly and regularly in the future because of climate and Akaki-Kality (0.40). Addis have moderate adaptive change (Nasiri and Shahmohammadi-Kalalagh 2013). capacity. Arada, Lideta,Yeka and Gulelle are sub cities Flood as exposure layer plays a key role in Addis Ababa, with low adaptive capacity. The adaptive capacity map of mainly in the southern and south eastern parts of the city Addis Ababa is presented on Fig. 7. due to the gentle slope characteristics of the relief in these The final vulnerability index, determined by composite parts. The past record on flooding in Addis Ababa indi- indicators of exposure, sensitivity and adaptive capacity in- cates, it was increasing from time to time. There are a wide dicates that sub cities with a normalized vulnerability index evidence that the flooding in Addis will be continue to in- value more than 0.5 were Arada(0.63), Addis Ketema(0.60), crease to the end of this century due to climate change Gulelle(0.53), Kirkos(0.52) and Lideta(0.5). Bole(0.45), Kolfe (CLUVA 2011; Ward and Lasage 2009; McSweeney et al. Keranio(0.4)and Nifas silk(0.37) were moderately vulner- 2010; Feyissa et al. 2018) and poor urban storm water man- able, while Yeka(0.33) and Akaki-Kality(0.30) were least agement in Addis Ababa. The exposure layer of climate Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 12 of 17 Fig. 6 Adaptive capacity layers (distance and road) change vulnerability identified in these analyses indicates Akaki-Kality area to be more vulnerable to flooding. Cli- the flood is common in poor infrastructure areas, low qual- mate change vulnerability activities and the vulnerability to ity housed but dense population along streams. Flooding is flooding is more aggravated due to a poor drainage system, common in Kirkos, parts of Bole and Akaki, at a time of rapid housing development along river banks and using heavy rain. The topographic nature of the southern and inappropriate construction material(Birhanu et al. 2016; south western are gentle, and the heavy rain drops from Belete 2011). The estimated cost of damage in Addis Ababa mountains flow to the southern direction, makes the was 373,640 million birr, 1.3 million birr and 1.3 million Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 13 of 17 Table 6 Adaptive Capacity: Access, distance and density Data Access Data Distance and Density Access to safe drinking water Access to sanitation Distance from ECC Distance from Health Centers Road Sub City (%) tap Norm (%) Of toilet Nor. Mean Dist. Norm. Mean Dist. Norm. Road Dens. Norm Arada 68.53 0.60 90.66 1.00 5.684 0.95 0.64 0.26 9.69 0.80 Addis Ketema 55.61 0.00 87.29 0.73 3.753 0.54 0.26 0.04 5.98 0.42 Lideta 65.91 0.48 87.75 0.77 2.293 0.23 0.89 0.41 3.10 0.12 Kirkos 73.19 0.82 90.31 0.97 3.090 0.40 0.25 0.03 11.58 1.00 Nifas silk 77.00 1.00 86.12 0.63 2.083 0.19 0.76 0.33 6.95 0.52 Akaki-K. 65.33 0.45 78.25 0.00 4.066 0.61 2.10 1.0 1.9 0.00 Bole 72.03 0.77 83.38 0.41 5.930 1.00 0.19 0.00 2.78 0.09 Yeka 71.65 0.75 84.97 0.54 1.368 0.04 2.06 1.09 4.46 0.26 Gulelle 73.11 0.82 84.74 0.52 4.688 0.74 0.25 0.03 8.60 0.89 Kolfe 68.64 0.61 84.98 0.54 1.197 0.00 2.10 1.0 6.81 0.51 % of tape inside the house, in compound private and tap in compound shared ECC emergency controlling center (Fire and Emergency Prevention and Rescue Agency) birr in 2010, 2011 and 2012 respectively. Many of the costs from the mud and earth used for the floors and walls of the recent flood damage, for instance flood damage oc- leads to increased susceptibility of the dwellers to curredin2006was notestimated.Majorityofthedamages respiratory diseases, especially among children (UN were occurred in the months of August and September as Habitat 2003). This is the reason why sub city like Bole well as in July. The stated Akaki-Kality and Kirkos were the is less sensitive than Addis Ketema and Arada. House- highest affected by flooding in Addis Ababa, due to their holds in slum areas usually occupy non-durable dwelling plain elevation and over-crowded of old houses. Bole and units that expose them to high morbidity and mortality Gullelle sub-cities had also high flood damage, while Addis risks (UN Habitat 2003). Exposure and sensitivity are al- Ketema Yeka, and Lideta are the moderately affected. Nefas most inseparable properties of a system (or community) silk Lafto, Arada and Kolfe Keraniyo sub-cities were the and are dependent on the interaction between the char- least affected sub cities(Fig. 8). acteristics of the system and on the attributes of the cli- The higher sensitivity of the sub-cities to climate mate stimulus. The exposure and sensitivity of a system change emanated from higher density, low quality of to an environmental related risk reflect the general con- constructed houses and low level of infrastructure devel- ditions and characteristics of the system. (Smit and opment. These areas are mainly found in the central Wandel 2006). The highest proportion of green area 2 2 parts of the city, which have a high population density coverage in Yeka(15.63km ) and Gulelle(15.63km )in mainly in Addis Ketema, Lideta and Arada sub cities. In line with their geographical position within the city, sensitivity values, variations were being observed among makes them to have lowest sensitivity to climate change. sub cities. Well planned, places with a good proportion In addition due historical reasons, the development of of green area coverage and places with good construc- botanic gardens in this area, the rehabilitation of forest tion materials were less sensitive to climate change is high. impacts. The predominant use of mud and wood for the The low level of socio-economic, demographic and construction of house walls and floors calls for frequent access to facilities in Addis Ababa made the adaptive repairs, which tend to be expensive in the long run. The capacity, to be low. Literacy rate and under five mortal- households that have the highest use of these materials ity rate as well as the higher unemployment rate also are in Akaki-Kality and Addis Ketema where the highest indicated the low level of the community’s adaptation need for repairs was also apparent. In addition, dust capacity to climate change. The low level of access layers, to social services like toilets and tap water were also the lowest. Another important layer in adaptive Table 7 Weights of Adaptive capacity layers capacity, which has a greatest influence on the adaptive Indicators 123456789 capacity layer, is the distance from the disaster control- Weight 0.11 0.12 0.11 0.14 0.13 0.12 0.10 0.08 0.09 ling center. Most parts, mainly the peripheral sub cities Total 1 had the highest distance from disaster controlling center. Literacy Rate(1),Under five Mortality Rate(2),Unemployment Rate(3),Activity The infrastructural development, used at the time of Rate(4), Access to tap(5), Access to toilets(6), Distance from emergency(7), Distance from health centers(8),Road Density(9), early warning and hazard was also an important factor; Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 14 of 17 Fig. 7 Adaptive Capacity Map which makes the adaptive capacity to be lower. The flooding as well. Lacks of adaptation to climate change sub-cities with have low adaptive capacities are well have increased vulnerability (Cochrane and Costolanski vulnerable to climate change. Due to rapid urbanization 2013). It is based on the adaptive capacity, whether any and population increase, low-income communities are type of climate-related impact occur, it is based on its forced to settle in flood-prone areas additionally the adaptive capacity whether it was a resilient or not. The poor drainage systems of the city also intensify the risk of vulnerability in Addis Ababa was exacerbated bylow level Table 8 Normalized Exposure, Sensitivity and Adaptive Capacity layers Exposure Sensitivity Adaptive Capacity 1 234 5678 910 11 12 13 14 15 Arada 0.92 1 0.22 0.75 1.00 0.77 0.00 0.33 0.68 0.51 0.60 1.00 0.95 0.26 0.80 Addis K. 1.0 0.87 0.39 1.00 1.00 1.00 0.77 0.11 1.00 0.35 0.00 0.73 0.54 0.04 0.42 Lideta 0.73 0.8 0.11 0.72 1.00 0.87 0.48 0.16 0.89 0.45 0.48 0.77 0.23 0.41 0.12 Kirkos 0.62 0.82 0.06 0.49 1.00 0.62 0.07 0.06 0.37 0.55 0.82 0.97 0.40 0.03 1.00 Nifas S. 0.23 0.18 0.11 0.14 0.99 0.31 0.57 0.01 0.35 0.40 1.00 0.63 0.19 0.33 0.52 Akaki K. 0.19 0.53 0.06 0.00 0.99 0.76 1.00 0.31 0.27 0.00 0.45 0.00 0.61 1.0 0.00 Bole 0.35 0.1 0.00 0.04 0.99 0.00 0.75 0 0.00 1.00 0.77 0.41 1.00 0.00 0.09 Yeka 0.00 0.13 0.32 0.10 0.00 0.82 0.48 1 0.60 0.32 0.75 0.54 0.04 1.09 0.26 Gulelle 0.31 0 1.00 0.25 0.43 0.97 0.62 0.42 0.33 0.32 0.82 0.52 0.74 0.03 0.89 Kolfe 0.23 0.1 0.44 0.19 0.84 0.62 0.81 0.27 0.61 0.43 0.61 0.54 0.00 1.0 0.51 SDV 0.34 0.39 0.30 0.34 0.34 0.77 0.32 0.29 0.30 0.25 0.28 0.29 0.36 0.35 0.37 1/SDV 2.96 2.57 3.37 2.92 2.94 1.00 3.15 3.39 3.30 4.00 3.60 3.46 2.79 2.86 2.68 C 0.02 0.02 0.02 0.02 0.02 0.87 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 Weight 0.06 0.05 0.07 0.06 0.06 0.62 0.07 0.07 0.07 0.09 0.08 0.07 0.06 0.06 0.06 Mean LST change(1), Air temperature change(2)Flood Risk(3), Population density(4), Forest cover(5), Mud(wood)house(6), Literacy Rate(7), Underfive Mortality Rate(8), Unemployment Rate(9), Activity Rate(10), Access to tap(11), Access to toilets(12), Distance from emergency(13), Distance fromhealth centers(14).Road Density(15) Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 15 of 17 Fig. 8 Frequency of flood by months (2010–2015) of economic development, low level of infrastructure, low The overall arrangement of the vulnerability in Addis level of access to basic needs, high rate of urbanization Ababa is determined by insufficient environmental man- and unimplemented environmental plans. agement capacity. Highest vulnerability is due to extreme The final vulnerability index, determined by composite precipitation events, which mainly affects densely popu- indicators of exposure, sensitivity and adaptive capacity lated areas with low level of infrastructure development indicates that sub cities with vulnerability index more than and low quality houses. Increases in temperature (both air 0.5 were Arada(0.63), Addis Ketema(0.60), Gulelle(0.53), and LST), as a result of urbanization activity and low land Kirkos(0.52) and Lideta(0.5). Bole(0.45), Kolfe Kera- use management, will also continue to be a key concern for nio(0.4)and Nifas silk(0.37) were moderately vulnerable Addis Ababa’s vulnerability to climate change in the future. subcities, while Yeka(0.33) and Akaki-Kality(0.30) were A change in these two climatic elements plays a key role in the least vulnerable sub cities based on the selected 15 determining the intensity and frequency of vulnerability. climate change vulnerability indicators. Well implemented planning activity was confirmed to play a pivotal role in reducing impacts exerted by climate change; the exposure and sensitivity to climate Conclusion change were confirmed to be high in densely populated This study tried to quantify, map and categorize climate areas with poor housing conditions,; temperature change change vulnerability in terms its sub contents of exposure, was confirmed to be high in sub cities found in the sensitivity and adaptive capacity. It considered 15 layers of northern and north western parts; the flooding risk was vulnerability indicators with bio physical, social and confirmed to be high in the areas of low slope and high economic layers at sub city level in Addis Ababa. The population density; and the highest vulnerability at a sub selections of the indicators are based on the literatures cities with lower adaptation capacities. The integrated used to identify and map climate change vulnerability Sullivan and Meigh’s Climate change vulnerability Index layers. Though it follows the IPCC’S climate change and the IPCC’`s definition of climate change vulnerabil- vulnerability analyses, the selection of the indicators were ity is well integrated in indicating and prioritizing based on the Sullivan and Meigh’s model which initially vulnerable hot spots to climate change at sub city level. developed to prepare climate change vulnerability index. Further studies have to be undertaken by adding add- Vulnerability to climate change is recognized as a state itional indicators and recent data for better understanding generated not just by climate change but by multiple and effective control of climate change vulnerability in processes and stressors. The stresses are more expressed Addis Ababa. by the low level development of adaptive capacity activ- ities in Addis Ababa. Changes in adaptive capacity which mainly contained the socio economic analyses were rap- Abbreviations idly changes, imposing a greatest effect on the impact of CLUVA: Climate Change Vulnerability Assessment in Africa; CSA: Central climate change vulnerability. Studies also suggest as the Statistical Agency; CVI: Climate Change Vulnerability Index; DEM: Digital Elevation Model; FAO: Food and Agriculture Organization; GCM: General changes in the social causes of vulnerability often Circulation Models; GDP: Gross Domestic Product; ICRISAT: International happen much more rapidly than many environmental Crops Research Institute for the Semi-Arid Tropics; IPCC: Intergovernmental changes(Khajuria and Ravindranath 2012.). Panel on Climate Change; LST: Land Surface Temperature Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 16 of 17 Acknowledgements change-and-vulnerability-of-African-cities-Research-briefs.pdf. Accessed Dec The authors would like to thank German Academic Exchange Service (DAAD) 2014. for providing in country scholarship to the corresponding author for his Ph.D Cochrane, L., and P. Costolanski. 2013. Climate change vulnerability and study at Addis Ababa University, Ethiopian Institute of Architecture, Building adaptability in an urban context: A case study of Addis Ababa, Ethiopia. Constriction and Urban Development. The authors would also like to thank International Journal of Sociology and Anthropology. 5 (6): 192–204. https:// Potsdam Institute for Climate Impact Research, for allowing a corresponding doi.org/10.5897/IJSA2013.0459. author a 6 month research visit in Potsdam Institute for Climate Impact Conway, D., and L.E.F. Schipper. 2011. Adaptation to climate change in Ethiopia: Research, Potsdam Germany. The USGS website that allowed the authors to Opportunities identified from Ethiopia. Global Environmental Change 21 (1): download the Landsat images freely from their archives should also be 227–237. https://doi.org/10.1016/j.gloenvcha.2010.07.013. acknowledged. The authors would also like to thank the Central statistical De Sherbinin, A. 2014. Climate change hotspots mapping: What have we Agency of Ethiopia for providing us population and socio economic data. learned? Climatic Change 123 (1): 23–37. https://doi.org/10.1007/s10584-013- 0900-7. Funding Doll, P. 2009. Vulnerability to the impact of climate change on renewable The first author is grateful to DAAD East Africa In country/In Region groundwater resources: A global-scale assessment. Environmental Research Scholarship funding programme number 57220758, for the PhD scholarship Letters 4 (1). https://doi.org/10.1088/1748-9326/4/3/035006. fund and Addis Ababa University, for Thematic Area Research Fund, fund Federal Democratic Republic of Ethiopia Central Statistical Agency. 2013. number TR/11/2013. Population projection of Ethiopia for all regions at Wereda level from 2014– 2017. https://www.academia.edu/30252151/Federal_Democratic_Republic_ of_Ethiopia_Central_Statistical_Agency_Population_Projection_of_Ethiopia_ Availability of data and materials for_All_Regions_At_Wereda_Level_from_2014_2017. Accessed Mar 2017. Authors declare that the data and materials presented in this manuscript can Feyissa, G., G. Zeleke, W. Bewket, and E. Gebremariam. 2018. Downscaling of be made publically available by Springer Open as per the editorial policy. future temperature and precipitation extremes in Addis Ababa under climate change. Climate. 6: 58. Authors’ contributions Folke, C., S. Carpenter, T. Elmqvist, L. Gunderson, C.S. Holling, and B. Walker. 2002. All authors read, helped to edit, and approved the final manuscript. Resilience and sustainable development: Building adaptive capacity in a world of transformations. A Journal of the Human Environment 31 (5): 437– Competing interests 440. https://doi.org/10.1579/0044-7447-31.5.437. The authors declare that they have no competing interests. Fussel, H.M., and J.T. Klein. 2006. Climate change vulnerability assessments. Climatic Change 75: 301–329 https://link.springer.com/article/10.1007/s10584- 006-0329-3. Publisher’sNote Gebreegziabher Z, Mekonnen A , Deribe, R, Boka, J, Abera S. 2016. Mapping Springer Nature remains neutral with regard to jurisdictional claims in vulnerability to climate change of the farming sector in the Nile Basin of published maps and institutional affiliations. Ethiopia. A micro-level perspective. Environment for development. http:// www.efdinitiative.org/sites/default/files/publications/efd-dp-16-19_0.pdf. Author details Accessed Aug 2017. Ethiopian Institute of Architecture, Building Construction and City Hassaan, M.A., M.A. Abdrabo, and P. Masabarakiza. 2017. GIS-based model for Development (EiABC), Environmental Planning Programme, Addis Ababa mapping malaria risk under climate change case study: Burundi. Journal of University, P.O. Box 518, Dej. Balcha Aba Nefso St, Addis Ababa, Ethiopia. Geoscience and Environment Protection 5: 102–117. https://doi.org/10.4236/ Ethiopian Institute of Architecture Building Construction and City gep.2017.511008. Development (EiABC), Addis Ababa University, Addis Ababa, Ethiopia. Water Intergovernmental Panel on Climate Change. 2014. Climate change 2014: and Land Resource Center, Addis Ababa, Ethiopia. Ethiopian Institute of Impacts, adaptation, and vulnerability WGII AR5 summary for policymakers. Architecture Building Construction and City Development (EiABC), Addis http://www.ipcc.ch/report/ar5/wg2/. Accessed Dec 2016. Ababa University, Addis Ababa, Ethiopia. Department of Geography and Environmental Studies, Addis Ababa University, Addis Ababa, Ethiopia. International Crops Research Institute for the Semi-Arid Tropics. 2009. Quantitative assessment of vulnerability to climate change from URL: http://danida.vnu.edu. Received: 9 May 2018 Accepted: 5 September 2018 vn/cpis/files/Papers_on_CC/Vulnerability/Quantitative%20assessment%20of%20 Vulnerability%20to%20Climate%20Change.pdf.Accessed Apr 2012. Iyengar, N.S., and P. Sudarshan. 1982. A method of classifying regions from multivariate data. Economic and Political Weekly 17: 2048–2052 http://www. References epw.in/journal/1982/51/special-articles/method-classifying-regions- Abebe, E., and A. Kebede. 2017. Assessment of climate change impacts on the multivariate-data.html. water resources of Megech River catchment, Abbay Basin. Ethiopia Open Khajuria, A., and N.H. Ravindranath. 2012. Climate change vulnerability Journal of Modern Hydrology. 07 (2). https://doi.org/10.4236/ojmh.2017.72008. assessment: Approaches DPSIR framework and vulnerability index. Journal of Adger, W.N., N.W. Arnell, and E.L. Tompkins. 2005. Successful adaptation to Earth Science and Climatic Change 3 (1). https://doi.org/10.4172/2157-7617. climate change across scales. Global Environmental Change 15 (2): 75–86. https://doi.org/10.1016/j.gloenvcha.2004.12.005. Legesse, S.A., P. Rao, and N. Rao. 2016. Statistical downscaling of daily Amare, A., and B. Simane. 2017. Determinants of smallholder farmers’ decision to temperature and rainfall data from global circulation models: In South adopt adaptation options to climate change and variability in the Muger sub Wollo zone, North Central Ethiopia. National Monthly Refereed Journal basin of the upper Blue Nile basin of Ethiopia. Agric & Food Secur 6 (64). of Res. in Science & Tech 2(7):27–39. https://doi.org/10.1186/s40066-017-0144-2. Malone, Battelle. 2009. Vulnerability and resilience in the face of climate change: Belete, D.H. 2011. Road and urban storm water drainage network integration in Current research and needs for population information. Washington: Pacific Addis Ababa: Addis Ketema sub-city. Journal of Engineering and Technology Northwest Division Richland http://www.globalchange.umd.edu/data/ Research 3 (7): 217–225 http://www.academicjournals.org/article/ publications/Resilience_and_Climate_Change.pdf. article1380190860_Belete.pdf. McSweeney C., New, M., Lizcano, G. 2010. UNDP climate change country profiles: Billi, P., Y. Alemu, and R. Ciampalini. 2015. Increased frequency of flash floods in Ethiopia. https://digital.library.unt.edu/ark:/67531/metadc226682/. Accessed Dire Dawa, Ethiopia: Change in rainfall intensity or human impact? Natural Aug 2015. Hazards 76 (2): 1373–1394. https://doi.org/10.1007/s11069-014-1554-0. Melese, S. 2016. Effect of climate change on water resources. Journal of Water Birhanu, D., H. Kima, C. Jang, and S. Park. 2016. Flood risk and vulnerability of Resources and Ocean Science. 5(1). https://doi.org/10.11648/j.wros.20160501.12. Addis Ababa City due to climate change and urbanization. Procedia Engineering 154: 696–702. https://doi.org/10.1016/j.proeng.2016.07.571. Moser, C., A. Norton, A. Stein, and S. Georgieva. 2010. Pro-poor adaptation to Climate change and Urban Vulnerability in Africa. 2011. Assessing vulnerability of climate change in urban Centres: Case studies of vulnerability and resilience urban systems. In Populations and goods in relation to natural and man-made in Kenya and Nicaragua. In Social development department report 54947-GLB. disasters in Africa http://www.cluva.eu/CLUVA_publications/CLUVA_Climate- Washington: World Bank. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 17 of 17 Moss, R.H., A.L. Brenkert, and E.L. Malone. 2001. Vulnerability to climate change: A Youneszadeh, S., N. Amiri, and P. Pilesjö. 2015. The effect of land use change on quantitative approach. Richland: Technical Report PNNL-SA-33642, Pacific land surface temperature in the Netherlands. The international archives of Northwest National Laboratories. the photogrammetry. Remote Sensing and Spatial Information Sciences XL-1/ Mwangi, K.M., and F. Mutua. 2015. Modeling Kenya’s vulnerability to climate change W5: 745–748. – A multifactor approach. International Journal of Science and Research (IJSR) Zurovec, H.G., S. Cadro, and B.K. Sitaula. 2017. Quantitative assessment of ISSN (Online):2319–7064 http://www.ijsr.net/archive/v4i6/SUB155079.pdf. vulnerability to climate change in rural municipalities of Bosnia and Accessed Sept 2015. Herzegovina. Sustainability 9 (7): 1208. https://doi.org/10.3390/su9071208. Nasiri, H., and S. Shahmohammadi-Kalalagh. 2013. Flood vulnerability index as a knowledge base for flood risk assessment in urban area. Journal of Novel Applied Science 2: 269–272. Olatona, O. 2017. Mapping of flood risk zones in Ala River basin Akure, Nigeria. American Journal of Applied Sciences and Engineering http://thescipub.com/ abstract/10.3844/ofsp.11774. Ouma, Y., and R. Tateishi. 2014. Urban flood vulnerability and risk mapping using integrated multi-parametric AHP and GIS: Methodological overview and case study assessment. Water Journal. https://doi.org/10.3390/w6061515. Rahmana, M. 2016. Land use and land cover changes and urban sprawl in Riyadh, Saudi Arabia: An analysis using multi-temporal Landsat data and Shannon's entropy index. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLI-B8: 1017–1021. https:// doi.org/10.5194/isprs-archives-XLI-B8-1017-2016. Shemsanga, C., A.N. Omambia, and Y. Gu. 2010. The cost of climate change in Tanzania: Impacts and adaptations. Journal of American Science 6 (3): 182–196 http://www.jofamericanscience.org/journals/am-sci/am0603/24_2189_ climate_am0603_182_196.pdf. Sierra-Soler, A., J. Adamowsk, Z. Qi, H. Saadat, and S. Pingale. 2015. High accuracy land use land cover (LULC) maps for detecting agricultural drought effects in Rainfed agro-ecosystems in Central Mexico. Journal of Water and Land Development and Environmental Engineering in Agriculture 26 (Vii–Ix): 19–35. https://doi.org/10.1515/jwld-2015-0014. Smit, B., and J. Wandel. 2006. Adaptation, adaptive capacity and vulnerability. Global Environmental Change 16: 282–292 http://citeseerx.ist.psu.edu/ viewdoc/download?doi=10.1.1.174.5145&rep=rep1&type=pdf. Srivastava, A. 2015. Report on the patterns of disaster risk reduction actions at local level. Vulnerability to climate change & variability: An investigation into macro & micro level assessments – A case study of agriculture sector in Himachal Pradesh, India. http://www.unisdr.org/campaign/resilientcities/ assets/documents/privatepages/ VULNERABILITY%20TO%20CLIMATE%20CHANGE%20&%20VARIABILITY.pdf. Sullivan, C., and J. Meigh. 2005. Targeting attention on local vulnerabilities using an integrated index approach: The example of the climate vulnerability index. Water Science and Technology 51 (5): 69–78 https://www.ncbi.nlm.nih. gov/pubmed/15918360. Tafesse, A., G. Ayele, M. Ketema, and E. Geta. 2016. Food security and vulnerability to climate change in eastern Ethiopia. Economics. 5 (6): 81–88. https://doi. org/10.11648/j.eco.20160506.11. Teferi, E., and H. Abraha. 2017. Urban Heat Island effect of Addis Ababa City: Implications of urban green spaces for climate change adaptation. Climate Change Adaptation in Africa, Climate Change Management. https://doi.org/10. 1007/978-3-319-49520-0_33. United Nations Development Programme. 2006. Human development report 2006. Beyond scarcity: Power, poverty and the global water crisis. http://hdr. undp.org/sites/default/files/reports/267/hdr06-complete.pdf. Assessed June United Nations Habitat. 2003. Urban inequities report. Addis Ababa, cities and citizens series. Ward P, and Lasage R. 2009. Downscaled climate change data from the HADCM3 and ECHAM5 models on precipitation and temperature for Ethiopia and Kenya. Institute for Environmental Studies, Vrije Universiteit. Netherlands. Woldegerima, T., K. Yeshitela, and S. Lindley. 2016. Characterizing the urban environment through urban morphology types (UMTs) mapping and land surface cover analysis: The case of Addis Ababa, Ethiopia. Urban Ecosystem. 19 (2): 245–263. https://doi.org/10.1007/s11252-016-0590-9. Woodruff, S., K. Vitro, and T. BenDor. 2017. GIS and coastal vulnerability to climate change, 236–257. Chapel Hill: University of North Carolina at Chapel Hill. https://doi.org/10.1016/B978-0-12-409548-9.09655-X. World Bank. 2015. Enhancing urban resilience Addis Ababa, Ethiopia. https:// www.gfdrr.org/sites/default/files/publication/Addis_Ababa_Resilient_cities_ program.pdf. Accessed 15 Sept 2016. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Geoenvironmental Disasters Springer Journals

GIS based quantification and mapping of climate change vulnerability hotspots in Addis Ababa

Download PDF
17 pages

Loading next page...
 
/lp/springer-journals/gis-based-quantification-and-mapping-of-climate-change-vulnerability-vZASvhDK3e

References (54)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Environment; Environment, general; Earth Sciences, general; Geography, general; Geoecology/Natural Processes; Natural Hazards; Environmental Science and Engineering
eISSN
2197-8670
DOI
10.1186/s40677-018-0106-4
Publisher site
See Article on Publisher Site

Abstract

Background: Unlike the studies undertaken on agricultural and hydrological sectors, focused climate change vulnerability researches in urban centers in Ethiopia is not widely available and of recent history. However, as many signals of climate change vulnerability started to happen in urban centers as well, it is inevitable to analyze, quantify, map, prioritize and be prepared for adaptation measures. This study is therefore, tried to assess, quantify and map climate change vulnerability in Addis Ababa, Ethiopia, by integrating two climate change vulnerability assessment models, namely, the Sullivan and Meigh’s Model of composite climate change vulnerability index and the IPCC’s approach of vulnerability assessment which comprises exposure, sensitivity and adaptive capacity. Fifteen sub- components of vulnerability indicators were identified in ten sub-cities (Addis Ketema, Arada, Akaki-Kality, Bole, Gulelle, Kirkos, Kolfe-Keraniyo, Lideta, Nifasilk-Lafto and Yeka) of Addis Ababa. Due to the scale, degree, amount and unit of measurement for the selected indicators varied, their values were normalized to a number which ranges between 0 and 1, indicating as the values increased to 1, vulnerability to climate change increases. The study uses Iyengar and Sudarshan’s unequal weighting system, to assign a weight to all indicators. The results were mapped using ArcGIS 10.2 package. Results: The results indicated that the ten sub-cities in Addis Ababa were found in different levels of vulnerability to climate change. The exposure and sensitivity were highest for Addis Ketema, Arada, and Lideta which are found in central parts of the city, with a normalization index value greater than 0.5. The adaptive capacity index is the highest in Gulelle, Bole, and Arada sub-cities. These sub-cities have better quality houses, well-planned districts, good infrastructural facilities and good coverage of green areas compared to others. The overall climate change vulnerability was the highest (normalized index > 0.5) in Arada, Addis Ketema and Lideta, due to the adaptation capacity is the lowest compared to other sub-cities. Conclusion: Addis Ababa is vulnerable to climate change impacts and the degree of vulnerability is underpinned by the interaction of multiple factors mainly adaptive capacities of sub-cities, location based characteristics and changes in climatic parameters. These present a need to strengthen mitigation and adaptation activities and prioritize sub cities for intervention based on the degree of vulnerability. It is also understood that the Sullivan and Meigh’sModel andIPCC’s approach for climate change analyses, could be used simultaneously for preparing vulnerability index per different geographical locations. Keywords: Climate change, Exposure, Sensitivity, Sullivanand Meighmodel,Vulnerability index * Correspondence: getyonani@yahoo.com Ethiopian Institute of Architecture, Building Construction and City Development (EiABC), Environmental Planning Programme, Addis Ababa University, P.O. Box 518, Dej. Balcha Aba Nefso St, Addis Ababa, Ethiopia Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 2 of 17 Background resilience reduces vulnerability (Folke et al. 2002). Vulner- Climate change is a major problem in the world in the ability to a climatic stressor is essentially a composite of ex- current century and it is the biggest and greatest challenge posure, a degree of sensitivity to the stressor, and the ability that humanity faced in the twenty-first Century. The recent of the exposed system to cope with the stressor. Vulner- fifth Assessment Report of the Intergovernmental Panel on ability is mostly considered to be a local phenomenon that Climate Change (IPCC 2014) indicated that each of the last is investigated by local-scale studies in which the specific three decades has been successively warmer at the Earth’s societal (and physical) situation can be taken into account surface than any preceding decade since 1850. The globally much better than in a global-scale study(Doll 2009). averaged combined land and ocean surface temperature Mapping vulnerability hotspots of climate change in data as calculated by a linear trend show a warming of 0.85 urban areas underpin climate change adaptation and miti- [0.65 to 1.06] °C over the period 1880 to 2012. It is well ac- gation policies and strategies in the contemporary world. It cepted that climate change will have a far more detrimental helps to figure out which places and people are the most effect on developing countries compared to developed vulnerable, as well as the degree of vulnerability and pos- countries; this is mainly because the capacity to respond to sible adaptation options (Gebreegziabher et al. 2016). Due such changes is the lowest in developing countries. More- to the potential of maps in depicting vulnerability hot spots over, it seems clear that vulnerability to climate change is is high, as it has strong visual elements and can easily be closely related to poverty, as the poor are least able to interpreted than text, its usage is increasing in different respond to climatic stimuli (Srivastava 2015). The case is research groups and organizations (de Sherbinin 2014). also true for Africa, as the continent is viewed as the most More specifically, vulnerability mapping helps to set vulnerable region to climate variability and change. Region- three policy measures. First, it is used to specify long-term ally in East Africa studies indicate that in countries like targets for the mitigation of climate change; second, to Burundi, Kenya, Sudan and Tanzania people are hit high identify vulnerable places and people and to prioritize re- by the impacts of climate change (Hassaan et al. 2017; source allocation for adaptation and to put adaptation Mwangi and Mutua 2015; Shemsanga et al. 2010). In measures (Fussel and Klein 2006). In climate change Ethiopia, within the last decades, high temperature values literature, two components of vulnerability are identified: recorded in different parts of the country. Various General internal and external dimensions (Gebreegziabher et al. Circulation Models (GCMs) results also tell as the future 2016). The internal dimension refers to frailty, insecurity, change climatic elements will be high. For instance, the and capacity to anticipate, cope with, resist and recover average future change for the whole of Ethiopia for three from adverse effects of shocks. The external dimension 30-year periods with A2 emissions shows warming in all involves exposure to risks and shocks. four seasons in all regions, with annual warming in In Ethiopia, the impact of climate change is studied Ethiopia by the 2020s of 1.2 °C with a range of 0.7–2.3 °C, largely at country level in different sectors mainly em- 2050s by 2.2 °C, range 1.4–2.9 °C (Conway and Schipper phasizing on the impacts of agricultural productivity and 2011). So far some studies predict long-term future climate water sectors. These studies reveal that climate change change situations that could prevail up to the end of this and variability severely affect economic conditions liveli- centuryinEthiopia(McSweeney etal. 2010). hoods and agricultural production and food security, If the occurrence of climate change is real, it is a must to water resources and health (Gebregiziabhre et al. 2016, identify it’s possible impacts and prepared for better adap- Legese et al. 2016, Tafesse et al. 2016, Melese 2016, tation mechanisms. In view of this, it is important to Abebe and Kebede 2017, Amare and Simane 2017). The understand the concept of climate change vulnerability, impact of climate change in urban scale is also studied identify indicators, quantify and map in order to have a in towns like Dire Dawa (Billi et al. 2015), Addis Ababa better understanding and to implement quick interven- (CLUVA 2011; Cochrane and Costolanski 2013; Birhanu tions. Climate change is not a single concept by itself and et al. 2016; Feyissa et al. 2018) indicating the identified rather it enclosed various concepts within it. In most cases, cities are vulnerable to climate change impacts mainly vulnerability, resilience, and adaptive capacity are the flooding and urban heat. central concepts for the analyses and understanding of the Though these studies provide some useful insights in impacts of climate change, because, together they provide a the area of vulnerability to climate change, and reflect framework that links biophysical climate sensitivity to the current efforts to understand the relationship social and economic factors that mitigate or amplify the between climate change and vulnerability, their scale of consequences of environmental changes. Vulnerability as- study is either at national level which is very coarse or at sessment and resilience is mainly analyzed with the topics case-specific level which do not indicate the overall like sustainability, hazards and climate change impacts vulnerability conditions. This study is, therefore analyzes (Malone 2009). Low resilience systems are intrinsically vul- how compiled bio-physical and socio-economic as well nerable to stress and shock, so in this sense increasing as geographic factors determine the vulnerability to Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 3 of 17 climate change in Addis Ababa, under different sub ad- its strategic role within the overall economic develop- ministration level (sub-cities). It analyses and construct ment of the country (World Bank 2015). vulnerabilty index at sub-city level, identify and map vul- The city is located in the central highlands of Ethiopia nerability hotspots. It also tries to know whether the vul- covering an areal extent of about 527 km with an aver- nerability to climate change is spatially varied or not in age elevation of 2600 m above mean sea level (asl). The one large city. In this regard, there is no prior studies altitude range extends from the highest peak at Mount available which integrates the IPCC’s approach and Entoto which is 3041 m to 2051 mean above sea level at Sullivan and Meigh’s model for vulnerability assessmnet the lower part of Akaki plain. The topography is undu- in Addis Ababa. The result of the study helps as an in- lating and form plateau in the northern, western and put for planners and different organizations which southwestern parts of the city. Bole, Akaki, and south already starts imposing a number of policies, strategies western part of the city is characterized by gentle and programs aimed at enhancing the adaptive capacity morphology and flat land areas. An average maximum and reducing the vulnerability of the country to climate temperature for Addis Ababa in the last 60 years is variability and change. 22.9 °C and the average minimum temperature is 10.2 °C. The average annual rainfall for Addis Ababa is 1184 mm. The wet season is from June to mid-September. The Methods urban area is endowed with three major rivers: Kebena, Study area description Little Akaki and Big Akaki rivers and numerous small Addis Ababa is found between 8°50′ N to 9°5′ N and streams. The population density varies from sub-city to 38°38′ E to 38°54′ E. It is the capital and the largest city sub-city. The highest density in Addis Ketema sub city of Ethiopia, with a total projected population of 3.44 (37,215p/sk.km) to the sparse density in Akaki-Kality million people in 2017(Federal Democratic Republic of sub-city (1832p/sk.km). All the sub-cities in the down- Ethiopia Central Statistical Agency 2013) and the admin- town have high population density compared to sub cit- istration of the city is divided into ten sub-cities. Addis ies found in peripheral areas. Fast rate of urban Ababa is home to 25% of the urban population in expansion and built up area are rapidly increasing in Ethiopia and is one of the fastest growing cities in Addis Ababa (Woldegerima et al. 2016), which were Africa. The city is a major growth corridor for the coun- largely attributed to population pressure on the land, a tries vision to become a middle income by 2025. The rapidly growing infrastructure and poor land use plan- city alone currently contributes approximately 50% to- ning (Teferi and Abraha 2017).The study area is pre- wards the national Gross Domestic Product, highlighting sented in Fig. 1. Fig. 1 Sub city boundary and Elevation map of Addis Ababa Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 4 of 17 Climate change vulnerability assessment approach climate change vulnerability. This study uses indicator ap- Various vulnerability assessment methods are available proach by integrating Climate change Vulnerability Index for urban areas. For instance, the Climate change and (CVI) model developed by Sullivan to construct climate Urban Vulnerability in Africa (CLUVA) project uses the change vulnerability index. This index is used as a checkup vulnerability assessment for Addis Ababa based on the in indicator approach, whereas the weight is assigned in works done by Moser et al. 2010, based on asset, institu- Arc GIS analyses. tion, attitude and physical as a collective umbrella of Vulnerability assessment follows the standardized vulnerability. work done by Sullivan and adopted by many types of According to the IPCC’s definition, vulnerability to research due to it comprises various physical and socio- climate change and variability is represented by three economic data for urban areas. The methodology used elements: exposure, sensitivity, and adaptive capacity. for CVI is based on the methodology of Water Poverty Exposure can be interpreted as the direct danger (i.e., Index developed by Sullivan and Meigh (2005) as speci- the stressor), and the nature and extent of changes to a fied in eq. 3. The main components and subcomponents region’s climate variables (e.g., temperature, precipita- of vulnerability index developed by Sullivan and Meigh tion, extreme weather events). Sensitivity describes the is presented in Table 1. human–environmental conditions that can worsen the w R þ w A þ w C þ w U þ w E þ w G hazard, ameliorate the hazard, or trigger an impact. r a c u e g CVI ¼ ð3Þ Adaptive capacity represents the potential to implement w þ w þ w þ w þ w þ w r a c u e g adaptation measures that help avert potential impacts. Adaptive capacity is considered “a function of wealth, R(Resources), A(Access), U(Use), C(Capacity), E(Envir- technology, education, information, skills, and infra- onment), G(Geospatial), and w,w ,w ,w ,w ,w – the r a u c e g structure, access to resources, and stability and manage- weights of indicators. ment capabilities” (Adger et al. 2005). However, slight modification can be done on the Therefore, analyzing vulnerability must involve identi- Sullivan model based on the actual context of the study fying not only the threat but also the “resilience,” or the area. Considering the basic indicators of climate change potential responsiveness of the system and its ability to exploit opportunities and resist or recover from the Table 1 Major Components of CVI (Based on Sullivan and negative effects of a changing environment. The first Meigh 2005) two components together represent the potential impact CVI Component Sub Indicators and adaptive capacity is the extent to which these im- Resource Total water resource pacts can be averted. Thus vulnerability is a potential Access Clean water impact (I) minus adaptive capacity (AC). Sanitation This leads to the following mathematical equation for Capacity Education level of a population vulnerability (International Crops Research Institute for Under five mortality rate the Semi-Arid Tropics 2009). Income V ¼ I‐AC ð1Þ People live in informal housing Which can be elaborated as follows (Moss et al. 2001): Access to a place of safety in the event of flooding Existence of disaster warning system Vulnerability ¼ðÞ adaptive capacity ‐ðÞ sensitivity þ exposure Use Domestic water use ð2Þ Industrial water use This study uses eqs. 1 and 2 which stated above, and Agricultural water use differentiates the indicators into three categories as Environment Water stress exposure, sensitivity and adaptive capacity. Water management Population density Climate change vulnerability index Loss of habitat Quantitative assessment of impact (vulnerability) is usually Geospatial Deforestation done by constructing a vulnerability index. This index is based on several sets of indicators that result in the vulner- Flood ability of a region. It produces a single number, which can Infrastructure be used to compare different regions (International Crops Thermal Heat Index Research Institute for the Semi-Arid Tropics 2009). Vari- Land conversion from natural vegetation ous indexes have been developed to construct and map Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 5 of 17 vulnerability with minor modifications in Sullivan and with vulnerability, the above formula will be changed to Meigh Model 2005, the methodology followed is integrat- the following as described in eq. 5. ing IPCC’s definition of vulnerability, in three sub layers Max X ‐X ij ij of exposure, sensitivity and adaptive capacity. Every com- X ¼ ð5Þ ij Max X ‐ Min X ponent is made up of sub-components. Based on these ij ij components, the 15 selected indicators presented on Once the normalized scores have been completed for Table 2 were identified and analyzed for Addis Ababa. entire selected indicators the simple average construct vulnerability index by adding all normalized scores (eq. Normalization of the indexes 6). Normalization of the indicators will undertake to ensure the P P comparability of the indicators. This was carried out using x þ y ij j j ij the methodology developed for the calculation of the Hu- VI ¼ ð6Þ man Development Index (UNDP 2006)using eq. 4.Then all the indicators assigned with values range from 0 to 1. Where VI is vulnerability index, i represent the sub-city and j represents indicator and K is the total number of indicators. X ‐ Min X ij ij X ¼ ð4Þ The value of the districts (i) is sub-cities and j (indicators) ij Max X ‐ Min X ij ij is replaced with normalized CVI results in this study. Where X is the separated value in the distribution, Assigning weights to indicators Min (X ) is the minimum value in the distribution and The basic challenge in constructing indices is the lack of ij Max (X ) is the maximum value of the mean of the dis- standard ways of assigning weight to each indicator. The ij tribution i is the sub-city and j is number of indicators. two most common weighting methods used to combine The value of the normalized equation falls between 0 indicators are equal and unequal weighting schemes (Geb- and 1. Where, 1 being the highest value and 0 with being reegziabher et al. 2016). The present study uses an unequal the least vulnerable area for the indicators with positive method of Iyengar and Sudarshan’s 1982 to give weight to relationship with climate changes vulnerability. Unless if all indicators. The choice of the weights in this manner the indicators are assumed to be negative relationship would ensure that large variation in any one of the Table 2 Selected Indicators of Exposure, Sensitivity and Adaptive capacity and their data source Component Indicators Functional Relation-ship Layers in Sullivan Definition Data Source Exposure Layers Mean air temperature change + Geospatial National Meteorological Agency Change in Land surface Temperature + Geospatial Extracted from Landsat Flood Risk + Geospatial Constructed from different layers Sensitivity Layers Mud(wood) house type + Capacity Central Statistical Agency Population density + Environment Central Statistical Agency Vegetation cover – Environment Extracted from 2017 Landsat Adaptive capacity Layers Unemployment rate + Capacity Central Statistical Agency Literacy rate – Capacity Central Statistical Agency Under five mortality rate + Capacity Central Statistical Agency Activity Rate Capacity Central Statistical Agency Distance from emergency centers + Capacity Addis Ababa City Fire & Emergency Prevention & Rescue Authority Road Length + Access Addis Ababa Road Authority Access to tap water + Access Central Statistical Agency Access to toilets + Access Central Statistical Agency Distance from all type of health – Access Integrated Land Information centers and Technology Office Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 6 of 17 indicators would not unduly dominate the contribution of Under temperature change layers two sub-layers have the rest of the indicators and distort the overall rankin- been prepared. One is air temperature change layer, g(Iyengar and Sudarshan 1982). which obtained from direct measurement of meteoro- In Iyengar and Sudarshan’s method, the weights are logical stations and the other is the Land surface assumed to vary inversely as the variance over the temperature change layer, which were derived from con- regions in the respective indicators of vulnerability. secutive Landsat satellite images. Regarding the air That is, the weight w is determined by eq. 7. temperature, three stations have been selected of five stations found in Addis Ababa based on the data avail- qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi W ¼ : ð7Þ ability and geographical variability. These were Entoto, Var X i ij Addis Ababa, and Bole observatories. The record at Entoto starts from 1989 onwards. Therefore the mean Where C is a normalizing constant: (Eq. (8)) temperature record from 1989 to 2016 was divided into 2 3 two parts for all the three stations. These are from −1 j¼k (1989–2002) and from (2004–2016). The mean values of qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 4 5 ½ C ¼ ð8Þ each station have been calculated and the first 14 years Var x j¼1 i ij average was subtracted from the first mean value. Based on this, the temperature values had obtained, mapped The overall sub city index, Yi,alsovariesfrom zero (0) and reclassified. The last final result was normalized. to one (1), with 1 indicating maximum vulnerability and 0 According to this, changes in mean temperature were indicating no vulnerability at all. The higher the normal- 0.47 °C, 0.24 °C and 0.20 °C at Entoto, Addis Ababa ob- ized sub-city index, the more the level of vulnerability. servatory, and Bole respectively. The normalized scores The composite indicator for climate change vulnerability were reclassified into five groups equally distributed with factors (exposure, sensitivity and resilience) for the ith sub 0.2 normalized values range as presented in Fig. 2. Based city was obtained as: on the Iyengar and Sudarshan’s method of eq. 7, the attached weights are (0.3), (0.3) and (0.4) respectively for W Y ð9Þ i ij Land Surface Temperature Change, air temperature Where: Yi is the composite indicator of ith sub-city; change, and flood risk. The assumption was that the W is the weight for each indicator lies between 0 and 1; places with the highest temperature change are more ∑W 1 and Y is the normalized scores of indicators. vulnerable to climate change. Identified exposure layers j= ij are presented in Table 3. Zonal statistics The mean difference of LST ranges from 3.4 °C at The Zonal Statistics tool of ArcGIS was widely used in Addis Ketema to 0.8 °C at Yeka sub-city. The difference this study to assign values obtained from raster images between the 1986 and 2017 has been normalized. Both to the sub-cities. A mean value of each indicator was results from air and land surface temperature change in- computed for every zone (sub city). The average of the dicates that high rate of change is found in the northern values in each zone is assigned to all output cells in the and northwestern parts of Addis Ababa while the low same zone. Zonal statistics tool is used in classifying and elevated parts of Addis Ababa, which are sparsely popu- relating land surface in land use change in various lated have low rate and magnitude of change in studies (Youneszadeh et al. 2015; Rahmana 2016; temperature. Sierra-Soler et al. 2015). Based on the calculations, the land surface temperature changes with time and the B). Flood Risk Layers. elevation, their statistics and spatial correlations has been utilized. The application of GIS to analyze climate The flood risk is not directly obtained as a single change vulnerability has grown exponentially in the last layer as explained in temperature case. Instead of con- decade (Woodruff et al. 2017). sidering a precipitation as a direct exposure layer in an urban area, it is better to integrate it with the flood risk Result layers. In order to get flood risk layers, four sub-layers Analyses of exposure components has been weighted using multi-criteria evaluation. Based on the climate vulnerability assessment procedure These four sub-layers are a) 2017 land cover layers ob- two major exposure layers were identified. These were tained from land cover layers. The land cover layer has temperature changes and flood risk layers. These layers five land cover classes, namely built-up area; bare are widely used in climate change exposure analyses. ground, open land, vegetation cover and agricultural land. b) Slope layers: the slope is divided into five clas- A). Temperature Layers. ses 0–2%, 2–8%, 8–15%, 15–30% and greater than 31%. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 7 of 17 Fig. 2 Normalized LST and air temperature change The slope layer is classified based on FAO, the highest which is found in the Southern and North East part of the slope value with fewer floods and the least with high city. Eutric Nitisol (111.55 km ) is the second most probability to be hit by flooding. c) Drainage density dominant soil found in the central and North West part layers: The drainage data is extracted from 30 m Digital of the region. Vertisol is characterized by fine textured Elevation Model (DEM of Addis Ababa. Up to six soil with > 60% clay in composition. As a result, the stream orders have been considered for the analyses. porosity of such soil is very fine making the movement 2 2 The drainage classes are 0–0.5 km/km ,0.5–1km/km , of material difficult within the soil. Hence, the perme- 2 2 2 1–1.5km ,1.5–2km and > 2 km/km . Drainage density ability of Vertisol is very low except within the cracks has a positive relationship with flooding. that are formed during dry seasons. d) The soil layers: As surveyed by Ministry of Water and Leptosol is characterized by shallow depth underlined Energy in 2004, the soil classes in Addis Ababa are Classic by hard rock and with less developed soil. The textural Xerosols, Chromic Luvisols, Eutric Nitisols, Leptosols, class is moderately coarse-textured soil with high perme- Orthic Solonchaks and Pellic Vertisols (Gizachew Kabite: ability (Gizachew Kabite: GIS and remote sensing based GIS and remote sensing based solid waste landfill site se- solid waste landfill site selection: A case of Addis Ababa lection: A case of Addis Ababa city, Unpublished). The city, Unpublished). Soils with high permeability are dominant soil of the region is Pellic Vertisol (277.23 km ) assigned lesser weights in flood risk mapping. Based on Table 3 Exposure Layers LST (°C) 1986 LST (°C) 2017 Mean Air Temperature (°C) a b Mean Mean Difference Nor. Mean Temp (°C) Mean Temp (°C) Mean Difference Nor. Arada 24.5 27.7 3.2 0.92 16.64 16.91 0.27 0.22 Addis Ketema 24.2 27.6 3.4 1.00 16.18 16.48 0.30 0.39 Lideta 25.2 27.9 2.7 0.73 17.05 17.30 0.25 0.11 Kirkos 25.3 27.7 2.4 0.62 17.06 17.30 0.24 0.06 Nifas Silk 26.6 28.0 1.4 0.23 16.65 16.90 0.25 0.11 Akaki-Kality 27.9 29.2 1.3 0.19 16.67 16.91 0.24 0.06 Bole 26.7 28.4 1.7 0.35 16.82 17.05 0.23 0.00 Yeka 25.1 25.9 0.8 0.00 16.09 16.38 0.29 0.32 Gulelle 22.4 24.0 1.6 0.31 14.04 14.45 0.41 1.00 Kolfe 25.3 26.7 1.4 0.23 15.92 16.23 0.31 0.44 Averaged mean air temperature record from (1989–2002) interpolated from Entoto, Addis Ababa, and Bole observatories, and derived for each sub city using Zonal Statistics Averaged mean air temperature record from (2003–2016) interpolated from Entoto, Addis Ababa, and Bole observatories, and derived for each sub city using Zonal Statistics (Source: Based on data from Landsat and National Meteorological Service, constructed by authors) Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 8 of 17 Fig. 3 Layers weighted for flood risk these the highest value is assigned for Vertisol and the flood risk.The food risk map is presented in a system- least value is assigned for Leptisol. In general 30%, 25%, atic manner from high risk to very low risk as depicted 30%, and 15% weight factor is assigned for land cover, in Fig. 4. The same approach is used for flood risk map- slope, drainage density, and soil type respectively to get ping by various authors(Olatona 2017;Ouma and flood risk map of Addis Ababa. Flood layers are Tateishi 2014). presented in Fig. 3. General exposure index indicates that the highest The final flood risk map was divided into five layers, exposure were in Addis Ketema(0.72), Arada(0.66) and from low to very high flood risk categories. The area Lideta(0.5). Moderate exposure were in Kirkos(0.45), and coverage percentage of very high risk and the high-risk Gulelle(0.49).Others sub-cities have low exposure index were summed together to categorize the sub-cities. value. The least exposure was in Bole(0.135). Based on the criteria, the sub-cities fall under risk areas The flood hazard map prepared from the composite were Arada, Lideta, Addis Ketma, Kirkos and Akaki, layers of drainage density, soil, population density and which have large parts of their land fall under flood risk. slope at the exposure layer of this analysis was well inte- The final flood risk map showed that areas with highest grated with the actual data recorded on flood history. population density and lowest elevation as well as high The flood history has been collected from the Addis drainage density have a highly exposed to climate Ababa fire and emergency preparedness office from change. The upper parts of the city which is covered by 2010 to 2016. The GPS points where flood history is forest had low flood risk while the places along the recorded in the past 8 years has been taken and overlaid drainage basins and high-density areas had highest as indicated in Fig. 4. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 9 of 17 Fig. 4 Flood risk map overlaid by frequency of flood Hazard records in Addis Ababa (Source:Data from Addis Ababa Fire and Emergency Service, compiled and constructed by authors) Sensitivity components sensitive, moderately sensitive and less sensitive areas have Three layers have been identified for sensitivity compo- been identified. The sensitivity is low in sparsely populated nent based on the criteria set by different authors and areas, in open spaces and for buildings constructed by a literature. These were the population density, the house standard material. The population density, the area of type, and the vegetation cover. Similar indicators have places covered by forest and the house type share per sub been used for sensitivity analyses (Gebreegziabher et al. city is presented in Table 4. 2016; Zurovec et al. 2017). High population density was The attached weights for each the three sensitivity layers at Addis Ketema(37,215/km ) and sparsely populated at were 0.32, 0.32 and 0.35 respectively for population density, Akaki-Kality(1832) in 2017 as projected by Central Stat- forest coverage and houses constructed by mud and wood istical Authority. Regarding the house types, the housing (Fig. 5). and population census have put different house types which Based on the weights, the highest sensitivity was in include corrugated iron sheets, concrete/cement, thatch, Addis ketema, Arada and Lideta,with very high sensi- wood and mud, bamboo, plastic, asbestos, and others. For tivity index greater than 0.8, high sensitivity index is in the analyses, only house types constructed of mud and Kirkos(0.69), the medium sensitivity is in Nefassilk wood have been selected. Most parts of Addis Ababa had Lafto, Kolfe, Akaki-Kality, and Gulelle with sensitivity the houses constructed from mud and wood. Arada, Addis index from 0.4 to 0.6. Yeka and Bole had a lowest sensi- Ketema, Kirkos are the sub-cities with the majority of their tivity index(0.2–0.4). houses are constructed by mud and wood. A well-planned sub city like in Bole sub-city, the main house constriction Adaptive capacity components type is concrete. It is assumed that houses constructed by Nine adaptive capacity layers have been identified. These mud are highly sensitive to climate change and can’tresist adaptive capacity layers were categorized into three change-induced impacts like heavy rainfall than concrete sub-layers. These are socio-economic and demographic building houses. The third subcomponent of sensitivity layers, access layers, and density and distance layers. a) layer is forest cover layer. It is extracted from 2017 Landsat The socio-economic layer consists of the unemployment image. The land cover analyses indicates that Yeka and rate, activity rate, literacy rate and under-five mortality Gulelle sub cities had a high area of forest cover while rates. The data were used based on the 2007 Ethiopian Addis Ketema, Arada, Kirkos, and Lideta have a small housing and population census’s definition. The proportion of vegetation covers. Places covered by forests percentage of socio-economic and demographic layers, are less sensitive to climate change. Based on these three per sub-cities, is presented in Table 5. layers one final layer is obtained. Accordingly highly b) Access to Social Services. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 10 of 17 Table 4 Sensitivity Layers Sub City Population Density (2017 projected) Forest Area(sq.km) Houses Constructed by Wood and Mud Pop density Normalized (Km ) Normalized (−ve relation) Percent Normalized Arada 28,206 0.75 0.08 1.00 81.2 0.77 Addis Ketema 37,215 1.00 0.07 1.00 88.4 1.00 Lideta 27,483 0.72 0.02 1.00 84.3 0.87 Kirkos 18,996 0.49 0.04 1.00 76.5 0.62 Nifas Silk 6812 0.14 0.22 0.99 66.5 0.31 Akaki-Kality 1832 0.00 0.42 0.99 80.7 0.76 Bole 3283 0.04 0.41 0.99 56.7 0.00 Yeka 5292 0.10 27.35 0.00 82.7 0.82 Gulelle 10,751 0.25 15.63 0.43 87.5 0.97 Kolfe 8479 0.19 4.43 0.84 76.3 0.62 Fig. 5 Sensitivity Layers Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 11 of 17 Table 5 Percentage and normalized values of socio economic and demographic data Sub City Literacy Under five Mortality Unemployment Activity Rate (%) Normalized Rate (%) Normalized Rate (%) Normalized Rate (%) Normalized Arada 89.3 0.00 0.05 0.33 24.2 0.68 63.8 0.51 Addis Ketema 84.0 0.77 0.03 0.11 27.2 1.00 62.8 0.35 Lideta 86.0 0.48 0.04 0.16 26.2 0.89 63.4 0.45 Kirkos 88.8 0.07 0.03 0.06 21.2 0.37 64.1 0.55 Nifas silk 85.4 0.57 0.02 0.01 21 0.35 63.1 0.40 Akaki-Kality 82.4 1.00 0.05 0.31 20.3 0.27 60.5 0.00 Bole 84.1 0.75 0.02 0 17.7 0.00 67 1.00 Yeka 86.0 0.48 0.10 1 23.4 0.60 62.6 0.32 Gulelle 85.0 0.62 0.05 0.42 20.8 0.33 62.6 0.32 Kolfe 83.7 0.81 0.04 0.27 23.5 0.61 63.3 0.43 There are many access data that has to be considered for vulnerable sub cities based on the 15 climate change climate change vulnerability. However, in order to avoid vulnerability indicators(Table 8). bias, only two data, which are fully available and complete for all sub cities were considered to analyze access. These Discussions are a percentage of access to water (only access to water Climate change impact is not measured only by the ex- provided by tap both in their house or compound whether posure and sensitivity’s strength; rather, it is a matter of it is shared or private) and percentage of access to toilets adaptive capacity. The exposure layers selected for vulner- for sanitation layer. c) Distance and Density Data: The two ability analyses in Addis Ababa was lower, but the sensitiv- layers of distance (distance from the health center and dis- ity was high and adaptive capacity activities were low. tance from emergency controlling center) have been These altogether makes Addis Ababa to be vulnerable city analyzed. The health centers include both private and gov- to climate change. The distribution of exposure layers ernment health centers, all types of clinics and all types of which contains physical factors was different. For instance hospitals. The distance from emergency controlling centers the air and LST temperature changes to the central and data indicates that the average distance of each place in the northern part of the city is high, indicating the higher of sub-city from the Addis Ababa fire and emergency control- the sensitivity and the lower value of adaptive capacity in ling center. The road density layer data was obtained by that parts. Addis Ketema and Arada sub cities, which con- adding all types of road lengths and dividing it to the area tained the oldest buildings within them, have old roofed of the sub-city. The types of roads considered in these houses, limited green area coverage’s and poorly managed analyses are asphalt, large stone, gravel, earth, and cobble streets with no street trees. These all makes the exposure stone. Thedistanceand densitylayersare presentedin to climate change in this part to be high. Fig. 6. Details of the adaptive capacity data with their In contrast to this, the sub-cities with high rise building, normalized value per sub-cities are given in Table 6. but with more planned parcels of land like in Bole and Finally as depicted on Table 7, the weights has been most parts of Yeka had lower-temperature change. Another assigned to the indicators. important and influential exposure indicator which has The final adaptive capacity index indicates Bole(0.56) high factor in Addis Ababa was flooding. Floods, the most was the highest sub city in adaptive capacity. Kolfe(0.48), prevalent of natural risks, are anticipated to happen more Addis Ketema(0.43), Kirkos(0.43) and Nifas Silk(0.41) strictly and regularly in the future because of climate and Akaki-Kality (0.40). Addis have moderate adaptive change (Nasiri and Shahmohammadi-Kalalagh 2013). capacity. Arada, Lideta,Yeka and Gulelle are sub cities Flood as exposure layer plays a key role in Addis Ababa, with low adaptive capacity. The adaptive capacity map of mainly in the southern and south eastern parts of the city Addis Ababa is presented on Fig. 7. due to the gentle slope characteristics of the relief in these The final vulnerability index, determined by composite parts. The past record on flooding in Addis Ababa indi- indicators of exposure, sensitivity and adaptive capacity in- cates, it was increasing from time to time. There are a wide dicates that sub cities with a normalized vulnerability index evidence that the flooding in Addis will be continue to in- value more than 0.5 were Arada(0.63), Addis Ketema(0.60), crease to the end of this century due to climate change Gulelle(0.53), Kirkos(0.52) and Lideta(0.5). Bole(0.45), Kolfe (CLUVA 2011; Ward and Lasage 2009; McSweeney et al. Keranio(0.4)and Nifas silk(0.37) were moderately vulner- 2010; Feyissa et al. 2018) and poor urban storm water man- able, while Yeka(0.33) and Akaki-Kality(0.30) were least agement in Addis Ababa. The exposure layer of climate Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 12 of 17 Fig. 6 Adaptive capacity layers (distance and road) change vulnerability identified in these analyses indicates Akaki-Kality area to be more vulnerable to flooding. Cli- the flood is common in poor infrastructure areas, low qual- mate change vulnerability activities and the vulnerability to ity housed but dense population along streams. Flooding is flooding is more aggravated due to a poor drainage system, common in Kirkos, parts of Bole and Akaki, at a time of rapid housing development along river banks and using heavy rain. The topographic nature of the southern and inappropriate construction material(Birhanu et al. 2016; south western are gentle, and the heavy rain drops from Belete 2011). The estimated cost of damage in Addis Ababa mountains flow to the southern direction, makes the was 373,640 million birr, 1.3 million birr and 1.3 million Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 13 of 17 Table 6 Adaptive Capacity: Access, distance and density Data Access Data Distance and Density Access to safe drinking water Access to sanitation Distance from ECC Distance from Health Centers Road Sub City (%) tap Norm (%) Of toilet Nor. Mean Dist. Norm. Mean Dist. Norm. Road Dens. Norm Arada 68.53 0.60 90.66 1.00 5.684 0.95 0.64 0.26 9.69 0.80 Addis Ketema 55.61 0.00 87.29 0.73 3.753 0.54 0.26 0.04 5.98 0.42 Lideta 65.91 0.48 87.75 0.77 2.293 0.23 0.89 0.41 3.10 0.12 Kirkos 73.19 0.82 90.31 0.97 3.090 0.40 0.25 0.03 11.58 1.00 Nifas silk 77.00 1.00 86.12 0.63 2.083 0.19 0.76 0.33 6.95 0.52 Akaki-K. 65.33 0.45 78.25 0.00 4.066 0.61 2.10 1.0 1.9 0.00 Bole 72.03 0.77 83.38 0.41 5.930 1.00 0.19 0.00 2.78 0.09 Yeka 71.65 0.75 84.97 0.54 1.368 0.04 2.06 1.09 4.46 0.26 Gulelle 73.11 0.82 84.74 0.52 4.688 0.74 0.25 0.03 8.60 0.89 Kolfe 68.64 0.61 84.98 0.54 1.197 0.00 2.10 1.0 6.81 0.51 % of tape inside the house, in compound private and tap in compound shared ECC emergency controlling center (Fire and Emergency Prevention and Rescue Agency) birr in 2010, 2011 and 2012 respectively. Many of the costs from the mud and earth used for the floors and walls of the recent flood damage, for instance flood damage oc- leads to increased susceptibility of the dwellers to curredin2006was notestimated.Majorityofthedamages respiratory diseases, especially among children (UN were occurred in the months of August and September as Habitat 2003). This is the reason why sub city like Bole well as in July. The stated Akaki-Kality and Kirkos were the is less sensitive than Addis Ketema and Arada. House- highest affected by flooding in Addis Ababa, due to their holds in slum areas usually occupy non-durable dwelling plain elevation and over-crowded of old houses. Bole and units that expose them to high morbidity and mortality Gullelle sub-cities had also high flood damage, while Addis risks (UN Habitat 2003). Exposure and sensitivity are al- Ketema Yeka, and Lideta are the moderately affected. Nefas most inseparable properties of a system (or community) silk Lafto, Arada and Kolfe Keraniyo sub-cities were the and are dependent on the interaction between the char- least affected sub cities(Fig. 8). acteristics of the system and on the attributes of the cli- The higher sensitivity of the sub-cities to climate mate stimulus. The exposure and sensitivity of a system change emanated from higher density, low quality of to an environmental related risk reflect the general con- constructed houses and low level of infrastructure devel- ditions and characteristics of the system. (Smit and opment. These areas are mainly found in the central Wandel 2006). The highest proportion of green area 2 2 parts of the city, which have a high population density coverage in Yeka(15.63km ) and Gulelle(15.63km )in mainly in Addis Ketema, Lideta and Arada sub cities. In line with their geographical position within the city, sensitivity values, variations were being observed among makes them to have lowest sensitivity to climate change. sub cities. Well planned, places with a good proportion In addition due historical reasons, the development of of green area coverage and places with good construc- botanic gardens in this area, the rehabilitation of forest tion materials were less sensitive to climate change is high. impacts. The predominant use of mud and wood for the The low level of socio-economic, demographic and construction of house walls and floors calls for frequent access to facilities in Addis Ababa made the adaptive repairs, which tend to be expensive in the long run. The capacity, to be low. Literacy rate and under five mortal- households that have the highest use of these materials ity rate as well as the higher unemployment rate also are in Akaki-Kality and Addis Ketema where the highest indicated the low level of the community’s adaptation need for repairs was also apparent. In addition, dust capacity to climate change. The low level of access layers, to social services like toilets and tap water were also the lowest. Another important layer in adaptive Table 7 Weights of Adaptive capacity layers capacity, which has a greatest influence on the adaptive Indicators 123456789 capacity layer, is the distance from the disaster control- Weight 0.11 0.12 0.11 0.14 0.13 0.12 0.10 0.08 0.09 ling center. Most parts, mainly the peripheral sub cities Total 1 had the highest distance from disaster controlling center. Literacy Rate(1),Under five Mortality Rate(2),Unemployment Rate(3),Activity The infrastructural development, used at the time of Rate(4), Access to tap(5), Access to toilets(6), Distance from emergency(7), Distance from health centers(8),Road Density(9), early warning and hazard was also an important factor; Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 14 of 17 Fig. 7 Adaptive Capacity Map which makes the adaptive capacity to be lower. The flooding as well. Lacks of adaptation to climate change sub-cities with have low adaptive capacities are well have increased vulnerability (Cochrane and Costolanski vulnerable to climate change. Due to rapid urbanization 2013). It is based on the adaptive capacity, whether any and population increase, low-income communities are type of climate-related impact occur, it is based on its forced to settle in flood-prone areas additionally the adaptive capacity whether it was a resilient or not. The poor drainage systems of the city also intensify the risk of vulnerability in Addis Ababa was exacerbated bylow level Table 8 Normalized Exposure, Sensitivity and Adaptive Capacity layers Exposure Sensitivity Adaptive Capacity 1 234 5678 910 11 12 13 14 15 Arada 0.92 1 0.22 0.75 1.00 0.77 0.00 0.33 0.68 0.51 0.60 1.00 0.95 0.26 0.80 Addis K. 1.0 0.87 0.39 1.00 1.00 1.00 0.77 0.11 1.00 0.35 0.00 0.73 0.54 0.04 0.42 Lideta 0.73 0.8 0.11 0.72 1.00 0.87 0.48 0.16 0.89 0.45 0.48 0.77 0.23 0.41 0.12 Kirkos 0.62 0.82 0.06 0.49 1.00 0.62 0.07 0.06 0.37 0.55 0.82 0.97 0.40 0.03 1.00 Nifas S. 0.23 0.18 0.11 0.14 0.99 0.31 0.57 0.01 0.35 0.40 1.00 0.63 0.19 0.33 0.52 Akaki K. 0.19 0.53 0.06 0.00 0.99 0.76 1.00 0.31 0.27 0.00 0.45 0.00 0.61 1.0 0.00 Bole 0.35 0.1 0.00 0.04 0.99 0.00 0.75 0 0.00 1.00 0.77 0.41 1.00 0.00 0.09 Yeka 0.00 0.13 0.32 0.10 0.00 0.82 0.48 1 0.60 0.32 0.75 0.54 0.04 1.09 0.26 Gulelle 0.31 0 1.00 0.25 0.43 0.97 0.62 0.42 0.33 0.32 0.82 0.52 0.74 0.03 0.89 Kolfe 0.23 0.1 0.44 0.19 0.84 0.62 0.81 0.27 0.61 0.43 0.61 0.54 0.00 1.0 0.51 SDV 0.34 0.39 0.30 0.34 0.34 0.77 0.32 0.29 0.30 0.25 0.28 0.29 0.36 0.35 0.37 1/SDV 2.96 2.57 3.37 2.92 2.94 1.00 3.15 3.39 3.30 4.00 3.60 3.46 2.79 2.86 2.68 C 0.02 0.02 0.02 0.02 0.02 0.87 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 Weight 0.06 0.05 0.07 0.06 0.06 0.62 0.07 0.07 0.07 0.09 0.08 0.07 0.06 0.06 0.06 Mean LST change(1), Air temperature change(2)Flood Risk(3), Population density(4), Forest cover(5), Mud(wood)house(6), Literacy Rate(7), Underfive Mortality Rate(8), Unemployment Rate(9), Activity Rate(10), Access to tap(11), Access to toilets(12), Distance from emergency(13), Distance fromhealth centers(14).Road Density(15) Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 15 of 17 Fig. 8 Frequency of flood by months (2010–2015) of economic development, low level of infrastructure, low The overall arrangement of the vulnerability in Addis level of access to basic needs, high rate of urbanization Ababa is determined by insufficient environmental man- and unimplemented environmental plans. agement capacity. Highest vulnerability is due to extreme The final vulnerability index, determined by composite precipitation events, which mainly affects densely popu- indicators of exposure, sensitivity and adaptive capacity lated areas with low level of infrastructure development indicates that sub cities with vulnerability index more than and low quality houses. Increases in temperature (both air 0.5 were Arada(0.63), Addis Ketema(0.60), Gulelle(0.53), and LST), as a result of urbanization activity and low land Kirkos(0.52) and Lideta(0.5). Bole(0.45), Kolfe Kera- use management, will also continue to be a key concern for nio(0.4)and Nifas silk(0.37) were moderately vulnerable Addis Ababa’s vulnerability to climate change in the future. subcities, while Yeka(0.33) and Akaki-Kality(0.30) were A change in these two climatic elements plays a key role in the least vulnerable sub cities based on the selected 15 determining the intensity and frequency of vulnerability. climate change vulnerability indicators. Well implemented planning activity was confirmed to play a pivotal role in reducing impacts exerted by climate change; the exposure and sensitivity to climate Conclusion change were confirmed to be high in densely populated This study tried to quantify, map and categorize climate areas with poor housing conditions,; temperature change change vulnerability in terms its sub contents of exposure, was confirmed to be high in sub cities found in the sensitivity and adaptive capacity. It considered 15 layers of northern and north western parts; the flooding risk was vulnerability indicators with bio physical, social and confirmed to be high in the areas of low slope and high economic layers at sub city level in Addis Ababa. The population density; and the highest vulnerability at a sub selections of the indicators are based on the literatures cities with lower adaptation capacities. The integrated used to identify and map climate change vulnerability Sullivan and Meigh’s Climate change vulnerability Index layers. Though it follows the IPCC’S climate change and the IPCC’`s definition of climate change vulnerabil- vulnerability analyses, the selection of the indicators were ity is well integrated in indicating and prioritizing based on the Sullivan and Meigh’s model which initially vulnerable hot spots to climate change at sub city level. developed to prepare climate change vulnerability index. Further studies have to be undertaken by adding add- Vulnerability to climate change is recognized as a state itional indicators and recent data for better understanding generated not just by climate change but by multiple and effective control of climate change vulnerability in processes and stressors. The stresses are more expressed Addis Ababa. by the low level development of adaptive capacity activ- ities in Addis Ababa. Changes in adaptive capacity which mainly contained the socio economic analyses were rap- Abbreviations idly changes, imposing a greatest effect on the impact of CLUVA: Climate Change Vulnerability Assessment in Africa; CSA: Central climate change vulnerability. Studies also suggest as the Statistical Agency; CVI: Climate Change Vulnerability Index; DEM: Digital Elevation Model; FAO: Food and Agriculture Organization; GCM: General changes in the social causes of vulnerability often Circulation Models; GDP: Gross Domestic Product; ICRISAT: International happen much more rapidly than many environmental Crops Research Institute for the Semi-Arid Tropics; IPCC: Intergovernmental changes(Khajuria and Ravindranath 2012.). Panel on Climate Change; LST: Land Surface Temperature Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 16 of 17 Acknowledgements change-and-vulnerability-of-African-cities-Research-briefs.pdf. Accessed Dec The authors would like to thank German Academic Exchange Service (DAAD) 2014. for providing in country scholarship to the corresponding author for his Ph.D Cochrane, L., and P. Costolanski. 2013. Climate change vulnerability and study at Addis Ababa University, Ethiopian Institute of Architecture, Building adaptability in an urban context: A case study of Addis Ababa, Ethiopia. Constriction and Urban Development. The authors would also like to thank International Journal of Sociology and Anthropology. 5 (6): 192–204. https:// Potsdam Institute for Climate Impact Research, for allowing a corresponding doi.org/10.5897/IJSA2013.0459. author a 6 month research visit in Potsdam Institute for Climate Impact Conway, D., and L.E.F. Schipper. 2011. Adaptation to climate change in Ethiopia: Research, Potsdam Germany. The USGS website that allowed the authors to Opportunities identified from Ethiopia. Global Environmental Change 21 (1): download the Landsat images freely from their archives should also be 227–237. https://doi.org/10.1016/j.gloenvcha.2010.07.013. acknowledged. The authors would also like to thank the Central statistical De Sherbinin, A. 2014. Climate change hotspots mapping: What have we Agency of Ethiopia for providing us population and socio economic data. learned? Climatic Change 123 (1): 23–37. https://doi.org/10.1007/s10584-013- 0900-7. Funding Doll, P. 2009. Vulnerability to the impact of climate change on renewable The first author is grateful to DAAD East Africa In country/In Region groundwater resources: A global-scale assessment. Environmental Research Scholarship funding programme number 57220758, for the PhD scholarship Letters 4 (1). https://doi.org/10.1088/1748-9326/4/3/035006. fund and Addis Ababa University, for Thematic Area Research Fund, fund Federal Democratic Republic of Ethiopia Central Statistical Agency. 2013. number TR/11/2013. Population projection of Ethiopia for all regions at Wereda level from 2014– 2017. https://www.academia.edu/30252151/Federal_Democratic_Republic_ of_Ethiopia_Central_Statistical_Agency_Population_Projection_of_Ethiopia_ Availability of data and materials for_All_Regions_At_Wereda_Level_from_2014_2017. Accessed Mar 2017. Authors declare that the data and materials presented in this manuscript can Feyissa, G., G. Zeleke, W. Bewket, and E. Gebremariam. 2018. Downscaling of be made publically available by Springer Open as per the editorial policy. future temperature and precipitation extremes in Addis Ababa under climate change. Climate. 6: 58. Authors’ contributions Folke, C., S. Carpenter, T. Elmqvist, L. Gunderson, C.S. Holling, and B. Walker. 2002. All authors read, helped to edit, and approved the final manuscript. Resilience and sustainable development: Building adaptive capacity in a world of transformations. A Journal of the Human Environment 31 (5): 437– Competing interests 440. https://doi.org/10.1579/0044-7447-31.5.437. The authors declare that they have no competing interests. Fussel, H.M., and J.T. Klein. 2006. Climate change vulnerability assessments. Climatic Change 75: 301–329 https://link.springer.com/article/10.1007/s10584- 006-0329-3. Publisher’sNote Gebreegziabher Z, Mekonnen A , Deribe, R, Boka, J, Abera S. 2016. Mapping Springer Nature remains neutral with regard to jurisdictional claims in vulnerability to climate change of the farming sector in the Nile Basin of published maps and institutional affiliations. Ethiopia. A micro-level perspective. Environment for development. http:// www.efdinitiative.org/sites/default/files/publications/efd-dp-16-19_0.pdf. Author details Accessed Aug 2017. Ethiopian Institute of Architecture, Building Construction and City Hassaan, M.A., M.A. Abdrabo, and P. Masabarakiza. 2017. GIS-based model for Development (EiABC), Environmental Planning Programme, Addis Ababa mapping malaria risk under climate change case study: Burundi. Journal of University, P.O. Box 518, Dej. Balcha Aba Nefso St, Addis Ababa, Ethiopia. Geoscience and Environment Protection 5: 102–117. https://doi.org/10.4236/ Ethiopian Institute of Architecture Building Construction and City gep.2017.511008. Development (EiABC), Addis Ababa University, Addis Ababa, Ethiopia. Water Intergovernmental Panel on Climate Change. 2014. Climate change 2014: and Land Resource Center, Addis Ababa, Ethiopia. Ethiopian Institute of Impacts, adaptation, and vulnerability WGII AR5 summary for policymakers. Architecture Building Construction and City Development (EiABC), Addis http://www.ipcc.ch/report/ar5/wg2/. Accessed Dec 2016. Ababa University, Addis Ababa, Ethiopia. Department of Geography and Environmental Studies, Addis Ababa University, Addis Ababa, Ethiopia. International Crops Research Institute for the Semi-Arid Tropics. 2009. Quantitative assessment of vulnerability to climate change from URL: http://danida.vnu.edu. Received: 9 May 2018 Accepted: 5 September 2018 vn/cpis/files/Papers_on_CC/Vulnerability/Quantitative%20assessment%20of%20 Vulnerability%20to%20Climate%20Change.pdf.Accessed Apr 2012. Iyengar, N.S., and P. Sudarshan. 1982. A method of classifying regions from multivariate data. Economic and Political Weekly 17: 2048–2052 http://www. References epw.in/journal/1982/51/special-articles/method-classifying-regions- Abebe, E., and A. Kebede. 2017. Assessment of climate change impacts on the multivariate-data.html. water resources of Megech River catchment, Abbay Basin. Ethiopia Open Khajuria, A., and N.H. Ravindranath. 2012. Climate change vulnerability Journal of Modern Hydrology. 07 (2). https://doi.org/10.4236/ojmh.2017.72008. assessment: Approaches DPSIR framework and vulnerability index. Journal of Adger, W.N., N.W. Arnell, and E.L. Tompkins. 2005. Successful adaptation to Earth Science and Climatic Change 3 (1). https://doi.org/10.4172/2157-7617. climate change across scales. Global Environmental Change 15 (2): 75–86. https://doi.org/10.1016/j.gloenvcha.2004.12.005. Legesse, S.A., P. Rao, and N. Rao. 2016. Statistical downscaling of daily Amare, A., and B. Simane. 2017. Determinants of smallholder farmers’ decision to temperature and rainfall data from global circulation models: In South adopt adaptation options to climate change and variability in the Muger sub Wollo zone, North Central Ethiopia. National Monthly Refereed Journal basin of the upper Blue Nile basin of Ethiopia. Agric & Food Secur 6 (64). of Res. in Science & Tech 2(7):27–39. https://doi.org/10.1186/s40066-017-0144-2. Malone, Battelle. 2009. Vulnerability and resilience in the face of climate change: Belete, D.H. 2011. Road and urban storm water drainage network integration in Current research and needs for population information. Washington: Pacific Addis Ababa: Addis Ketema sub-city. Journal of Engineering and Technology Northwest Division Richland http://www.globalchange.umd.edu/data/ Research 3 (7): 217–225 http://www.academicjournals.org/article/ publications/Resilience_and_Climate_Change.pdf. article1380190860_Belete.pdf. McSweeney C., New, M., Lizcano, G. 2010. UNDP climate change country profiles: Billi, P., Y. Alemu, and R. Ciampalini. 2015. Increased frequency of flash floods in Ethiopia. https://digital.library.unt.edu/ark:/67531/metadc226682/. Accessed Dire Dawa, Ethiopia: Change in rainfall intensity or human impact? Natural Aug 2015. Hazards 76 (2): 1373–1394. https://doi.org/10.1007/s11069-014-1554-0. Melese, S. 2016. Effect of climate change on water resources. Journal of Water Birhanu, D., H. Kima, C. Jang, and S. Park. 2016. Flood risk and vulnerability of Resources and Ocean Science. 5(1). https://doi.org/10.11648/j.wros.20160501.12. Addis Ababa City due to climate change and urbanization. Procedia Engineering 154: 696–702. https://doi.org/10.1016/j.proeng.2016.07.571. Moser, C., A. Norton, A. Stein, and S. Georgieva. 2010. Pro-poor adaptation to Climate change and Urban Vulnerability in Africa. 2011. Assessing vulnerability of climate change in urban Centres: Case studies of vulnerability and resilience urban systems. In Populations and goods in relation to natural and man-made in Kenya and Nicaragua. In Social development department report 54947-GLB. disasters in Africa http://www.cluva.eu/CLUVA_publications/CLUVA_Climate- Washington: World Bank. Feyissa et al. Geoenvironmental Disasters (2018) 5:14 Page 17 of 17 Moss, R.H., A.L. Brenkert, and E.L. Malone. 2001. Vulnerability to climate change: A Youneszadeh, S., N. Amiri, and P. Pilesjö. 2015. The effect of land use change on quantitative approach. Richland: Technical Report PNNL-SA-33642, Pacific land surface temperature in the Netherlands. The international archives of Northwest National Laboratories. the photogrammetry. Remote Sensing and Spatial Information Sciences XL-1/ Mwangi, K.M., and F. Mutua. 2015. Modeling Kenya’s vulnerability to climate change W5: 745–748. – A multifactor approach. International Journal of Science and Research (IJSR) Zurovec, H.G., S. Cadro, and B.K. Sitaula. 2017. Quantitative assessment of ISSN (Online):2319–7064 http://www.ijsr.net/archive/v4i6/SUB155079.pdf. vulnerability to climate change in rural municipalities of Bosnia and Accessed Sept 2015. Herzegovina. Sustainability 9 (7): 1208. https://doi.org/10.3390/su9071208. Nasiri, H., and S. Shahmohammadi-Kalalagh. 2013. Flood vulnerability index as a knowledge base for flood risk assessment in urban area. Journal of Novel Applied Science 2: 269–272. Olatona, O. 2017. Mapping of flood risk zones in Ala River basin Akure, Nigeria. American Journal of Applied Sciences and Engineering http://thescipub.com/ abstract/10.3844/ofsp.11774. Ouma, Y., and R. Tateishi. 2014. Urban flood vulnerability and risk mapping using integrated multi-parametric AHP and GIS: Methodological overview and case study assessment. Water Journal. https://doi.org/10.3390/w6061515. Rahmana, M. 2016. Land use and land cover changes and urban sprawl in Riyadh, Saudi Arabia: An analysis using multi-temporal Landsat data and Shannon's entropy index. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLI-B8: 1017–1021. https:// doi.org/10.5194/isprs-archives-XLI-B8-1017-2016. Shemsanga, C., A.N. Omambia, and Y. Gu. 2010. The cost of climate change in Tanzania: Impacts and adaptations. Journal of American Science 6 (3): 182–196 http://www.jofamericanscience.org/journals/am-sci/am0603/24_2189_ climate_am0603_182_196.pdf. Sierra-Soler, A., J. Adamowsk, Z. Qi, H. Saadat, and S. Pingale. 2015. High accuracy land use land cover (LULC) maps for detecting agricultural drought effects in Rainfed agro-ecosystems in Central Mexico. Journal of Water and Land Development and Environmental Engineering in Agriculture 26 (Vii–Ix): 19–35. https://doi.org/10.1515/jwld-2015-0014. Smit, B., and J. Wandel. 2006. Adaptation, adaptive capacity and vulnerability. Global Environmental Change 16: 282–292 http://citeseerx.ist.psu.edu/ viewdoc/download?doi=10.1.1.174.5145&rep=rep1&type=pdf. Srivastava, A. 2015. Report on the patterns of disaster risk reduction actions at local level. Vulnerability to climate change & variability: An investigation into macro & micro level assessments – A case study of agriculture sector in Himachal Pradesh, India. http://www.unisdr.org/campaign/resilientcities/ assets/documents/privatepages/ VULNERABILITY%20TO%20CLIMATE%20CHANGE%20&%20VARIABILITY.pdf. Sullivan, C., and J. Meigh. 2005. Targeting attention on local vulnerabilities using an integrated index approach: The example of the climate vulnerability index. Water Science and Technology 51 (5): 69–78 https://www.ncbi.nlm.nih. gov/pubmed/15918360. Tafesse, A., G. Ayele, M. Ketema, and E. Geta. 2016. Food security and vulnerability to climate change in eastern Ethiopia. Economics. 5 (6): 81–88. https://doi. org/10.11648/j.eco.20160506.11. Teferi, E., and H. Abraha. 2017. Urban Heat Island effect of Addis Ababa City: Implications of urban green spaces for climate change adaptation. Climate Change Adaptation in Africa, Climate Change Management. https://doi.org/10. 1007/978-3-319-49520-0_33. United Nations Development Programme. 2006. Human development report 2006. Beyond scarcity: Power, poverty and the global water crisis. http://hdr. undp.org/sites/default/files/reports/267/hdr06-complete.pdf. Assessed June United Nations Habitat. 2003. Urban inequities report. Addis Ababa, cities and citizens series. Ward P, and Lasage R. 2009. Downscaled climate change data from the HADCM3 and ECHAM5 models on precipitation and temperature for Ethiopia and Kenya. Institute for Environmental Studies, Vrije Universiteit. Netherlands. Woldegerima, T., K. Yeshitela, and S. Lindley. 2016. Characterizing the urban environment through urban morphology types (UMTs) mapping and land surface cover analysis: The case of Addis Ababa, Ethiopia. Urban Ecosystem. 19 (2): 245–263. https://doi.org/10.1007/s11252-016-0590-9. Woodruff, S., K. Vitro, and T. BenDor. 2017. GIS and coastal vulnerability to climate change, 236–257. Chapel Hill: University of North Carolina at Chapel Hill. https://doi.org/10.1016/B978-0-12-409548-9.09655-X. World Bank. 2015. Enhancing urban resilience Addis Ababa, Ethiopia. https:// www.gfdrr.org/sites/default/files/publication/Addis_Ababa_Resilient_cities_ program.pdf. Accessed 15 Sept 2016.

Journal

Geoenvironmental Disasters – Springer Journals

Published: Sep 21, 2018

There are no references for this article.