|
PAGE 1
Economic Vulnerability and Disaster Risk Assessment in Malawi and MozambiqueMeasuring Economic Risks of Droughts and Floods THE WORLD BANK
PAGE 2
Highly variable climate in Malawi and Mozambique has a signicant inuence on the amount, timing, and frequency of precipitation events and runoff patterns, which results in frequent recurrent droughts and oods.
PAGE 3
| 1INTRODUCTION Highly variable climate in Malawi and Mozambique has a signicant inuence on the amount, timing, and frequency of precipitation events and runoff patterns, which results in frequent recurrent droughts and oods. In Malawi rainfall varies considerably both seasonally and from year to year. The country has one of the most erratic rainfall patterns in Africa. Between 1967 and 2003, the country experienced six major droughts, affecting over 21 million people in total. Floods occur in southern Malawi, particularly in the Lower Shire River valley and the lakeshore areas of Lake Malawi, Lake Malombe and Lake Chilwa, as well as in the lower reaches of the Songwe River in the northern region. Between 1967 and 2003, 18 oods were recorded killing at least 570 people, rendering 132,000 homeless, and affecting a total of 1.8 million people. Flooding also damages property and infrastructure, impeding drainage of agricultural lands and causing crop damage. In Mozambique, drought is the most frequent natural disaster, occuring every three to four years. Mozambique has areas that are classied as semi-arid and arid where raineven when above averageis inadequate, resulting in critical water shortage and limited agriculture productivity. It is estimated that droughts contributed to about 4,000 deaths between 1980 and 2000. A number of geographical factors cause oods in Mozambique, which give rise to high coefcients of rainfall variability. The country is situated downstream of nine major international river basins. It lies in the path of tropical cyclones formed in the Indian Ocean and is affected by three to four cyclones a year. In 2000, Mozambique experienced its worst oods in 50 years, killing about 100 people and displacing 540,000. The economies of Malawi and Mozambique are strongly natural resource dependent, and agriculture accounts for about one-third of their G D P. Major oods and droughts in Malawi and Mozambique have a signicant impact on national economic performance. Measures of uctuations in G D P and in growth rates of agricultural and non-agricultural sector products demonstrate the sensitivity of the economies to water shocks. T he expectation of variability and the unpredictability of rainfall and runoff constrain opportunities for growth by encouraging risk averse behavior and discouraging investments in land improvements, advanced technologies, and agricultural inputs. An unreliable water supply due to hydrological variability is a signicant disincentive to investments in industry and services, slowing the diversication of economic activities. Frequent droughts and oods pose a systematic risk to the economies of Malawi and Mozambique and require the development of a longer-term approach for adapting to drought and ood risks and to conditions of chronic hydrological variability. The objective of the Economic Vulnerability and Disaster Risk Assessment study was to assist the Governments of Malawi and Mozambique in determining the extent of their economic vulnerability to climate variability and in assessing the systematic economic risks associated with droughts and oods. The study results create a compelling argument for incorporating disaster risk reduction (DRR) into national economic planning and development strategies. T he study also quantied the depth and extent of hazard, vulnerability, and disaster loss associated with oods and droughts through detailed spatial risk proling for both countries. T he study results include the Atlases of D rought and Flood R isks for Malawi and Mozambique to inform various stakeholders and decision makers involved with disaster preparedness and response about the the parameters and distribution of drought and ood risks.
PAGE 4
2 | SAMOA Post Disaster Needs Assessment The study applied probabilistic risk analysis to evaluate the impacts of the natural hazards. The essential building blocks in the probabilistic risk analysis are hazard, exposure, vulnerability, and loss modules.
PAGE 5
Measuring Economic Risks of Droughts and Floods | 3STUDY METHODOLOGY The study applied probabilistic risk analysis to evaluate the impacts of the natural hazards. The essential building blocks in the probabilistic risk analysis are hazard, exposure, vulnerability, and loss modules. Hazard represents the occurrence and severity of adverse events. Exposure characterizes the asset(s) at risk. Vulnerability describes the potential damage to the exposure, corresponding to varying degrees of hazard severity. Risk is expressed in terms of the probability of exceeding specic levels of direct losses (in physical and monetary terms). Figure 1 illustrates the methodological framework adopted in this study. Figure 1. Drought and Flood Risk Modeling Framework Climate Change Scenarios Assessment of Vulnerabilty through Exposure Data Loss Estimates (in Kwacha) = Risk value x value of exposure Physical InfrastructureV = f {Housing (age, construction materials), Public facilities such as roads, bridges (age, types)}Physical InfrastructureV = f {Rainfall (SPI), Duration and Distribution) Risk Value = f {V, damage to hazard} Structural (Physical)Non-Structural (Biological) Crop Growth ModelV = f {Weather (rtm, tm, RH, sr),soil (texture, ph, depth), crop management practices (irrigation, fertilizer, seed, mechanization)} Planning services Department of Water resources Metheorological services National Statistics Ofce, NSOData Frequency and Severity of Hazard by Location Hydrological Meteorological Hazards FLOOD DROUGHT
PAGE 6
4 | Economic Vulnerability and Disaster Risk Assessment in Malawi and Mozambique The implementation of the methodology was subdivided into six modules: Stochastic Weather Module generates a set of stochastic events derived from the characteristics of historical weather events. Each stochastic weather event is quantied in terms of its location, severity, and rate of occurrence. Drought and Flood Hazard Module estimates the geographic impact of an event in terms of its severity, frequency, (drought duration and intensity, ood depth and extent), and probability of future occurrence. In the case of oods, for each event in the stochastic set, a footprint is generated that quanties the severity distribution over the area affected. The probability of the event occurrence within a given year is measured by an events return period (RP), which is the expected length of time between the reoccurrence of two events with similar characteristics. In this analysis, weather events were evaluated across the full spectrum of return periods. Exposure Module denes the spatial distribution of the asset(s)-at-risk and classies them based on the entailed potential damage to the relevant levels of hazards. In drought risk modeling, the principal assets at risk for Malawi are maize and tobacco and for Mozambique are maize and sorghum. The exposure parameters include the crop area and its production. In addition to the agricultural exposure, the residential (population and households) and infrastructural (roads and railways) exposures are considered in the study. Vulnerability Module quanties the vulnerability of the assets subjected to the hazard. In the case of droughts, biological vulnerability is assessed by modeling the potential losses in the crop yields due to droughts at different phenological stages of their crop cycle. The risk is expressed in terms of production losses. The study also developed drought indices, using statistical and deterministic agro-meteorological models to simulate responses to probabilistic crop losses. In the case of oods, the crop production losses are estimated by relating the crop area and production with seasonal maximum ows at various gauge sites. The physical vulnerability of the infrastructureroads and household dwellingsis computed using ood depth and extent with corresponding MDR (mean damage ratio). Direct Loss Computation Module calculates the total direct loss in monetary terms by combining the vulnerability with the value of the affected assets into nancial loss costs. Macroeconomic Module estimates economy-wide impacts of droughts and oods in Malawi by imposing estimated direct crop production losses on an economy-wide general equilibrium model (CGE). By integrating information from a wide range of sources, including national accounts, foreign trade data, and a 2004-2005 household survey, the model estimated macro impacts on agricultural and national GDP and micro impacts on employment and household poverty.
PAGE 7
Measuring Economic Risks of Droughts and Floods | 5 Assessing Drought RisksDrought occurs when the precipitation or soil moisture levels are sufciently below the long-run mean. In this study the Standard Precipitation Index (SPI), which is based on precipitation data, is used for drought identication. This index permits the measurement of drought intensity, magnitude, or severity, as well as its duration. The probability of an event occurring within a certain year is estimated on the basis of historical data. The measured SPI is adjusted to control for when the event took place during the growing cycle (i.e., November to March). Regression models are used to identify whether a statistical, non-linear relationship exists between historical drought events of different severities (i.e., as measured by their SPIs) and the associated crop production losses for different crops observed during those years. Production losses are calculated as the difference between observed production and expected production where the latter reects the production level achieved during the closest normal or non-drought year. The regression coefcients are then used in a stochastic model that randomly generates a large number of possible drought events across the full range of return periods (RPs). The relationship between different drought events and their associated production losses is dened and represented by a loss exceedance curve (LEC). In the context of agricultural risk, a LEC gives the likelihood or probability that a certain level of crop loss will be exceeded during a particular drought event. The average annual loss (AAL) and the probable maximum loss (PML) describe the expected long-term loss (Box 1). Box 1. Risk Metrics Loss Exceedance Curve (LEC): Using a probabilistic model, the probability for a certain level of loss to be exceeded during a particular (drought or ood) event is determined and represented by a LEC. The LECs allow attachment of a precise probability of occurrence to each possible weather event. Average Annual Loss (AAL): While future weather patterns are uncertain, expected long-term losses can be predicted with greater certainty. This expected long-term loss is the average annual loss (AAL), which is obtained by multiplying the probability of an event by its expected loss and summing over all possible events (i.e., integration of the LEC). Probable Maximum Loss (PML): This represents the largest possible loss that may occur in regard to the risk of the event. In this study, PML is dened as the loss that occurs in case of a ood or drought event of 100-year return period. The drought hazard and exposure (crop area and production) maps were simulated, highlighting the areas that have a 1-in-5, 10, 20, 50 and 100 year return period frequency. The generated risk maps spatially portray the drought impacts on crop area and production across the countries (Figure 2). D rought vulnerability is expressed in terms of mean damage ratio (M DR ), which represents the ratio of the loss in crop production to the total crop production under the normal weather conditions. T he M DR maps were generated corresponding to 1-in-5, 10, 20, and 50 year droughts (Figure 3). R isk is expressed as the probability of exceeding specic levels of physical and direct nancial losses. In the present context, drought risk is presented as production losses (tonnage) and equivalent monetary losses ( U S$). T hese are generated corresponding to 1-in-5, 10, 20, and 50 year droughts along with AAL (annual recurrence) and PML (1-in-100 year).
PAGE 8
6 | Economic Vulnerability and Disaster Risk Assessment in Malawi and Mozambique Figure 2. Drought Hazard Maps for Malawi and Mozambique for a 1-in-5 Y ear Return Period (RP). Figure 3. Mapping Drought Vulnerability for Malawi and Mozambique
PAGE 9
Financial and Insurance Instruments | 7Assessing Flood RisksThe ood risk model adopts a similar approach to the drought model in that hazard is assessed using estimates of the probability of occurrence for oods of different severities. The probabilistic risk model is based on runoff, which means that observed ood discharges are used to identify oods and to estimate their probability of occurrence. Stochastically generated discharges are then routed through a Digital Elevation Model of the affected oodplain to determine ood extents and depths at a detailed 90 meter resolution. The stochastic results from this model were validated using satellite images of historical ood events. Agricultural losses are determined on the basis of information about farmers exposure to ood events. This depends on the portion of cultivated land in geographic areas likely to be inundated during oods of different severities. As with the drought analysis, regression models are used to estimate the relationship between production levels and historical ood events. The physical vulnerability of the infrastructureroads and household dwellingsis computed using ood depth and extent with corresponding MDR (mean damage ratio). Data from the regression models were incorporated into a stochastic ood model in order to generate production losses under the complete distribution of ood events (i.e., for all RPs). Detailed spatial information on ood hazard and vulnerability is necessary for ood risk assessment and is extremely important for the preparedness and prevention strategies of governments, as well as for individual stakeholders, such as communities, enterprises, and house owners. The hazard module generates hazard intensities (in terms of ood depth) at the district-level resolution for each stochastic event. The vulnerability module relates ood depths with damage susceptibility to houses, infrastructures, and agricultural areas. The exposure module describes the inventory of different types of dwellings, roads, rail network, and agricultural areas at a district-level resolution. Figures 4 and 5 map illustrate some of the modeling results for Malawi and Mozambique. Figure 4. Flood Hazard and Exposure in the Shire Basin, Malawi
PAGE 10
8 | Economic Vulnerability and Disaster Risk Assessment in Malawi and Mozambique Long-Term Losses The loss module quanties the losses incurred by the assets as dened by the exposure module. Uncertainty over when droughts and oods will occur is a serious problem for short-term risk management. However, the Annual Average Loss (AAL) calculated based on the drought and ood LECs is an important characteristic of the systematic losses due droughts and oods in the long term. These calculations show that, for example, Malawi loses, on average, 4.6 percent of its maize production each year due to droughts (based on todays adoption of different varieties). Similarly, about 12 percent of maize production is lost each year to ooding in the southern region where about one-third of Malawis maize is grown. Droughts and oods are, therefore, a major obstacle for agriculture and food security in the country. Figures 6 and 7 summarize some results of the assessment of direct losses for Malawi and Mozambique. Figure 5. Flood Hazard and Exposure (Schools and Households) for 1-in-10 year ood in Mozambique
PAGE 11
Measuring Economic Risks of Droughts and Floods | 9 Figure 6. Malawi: Long-term Direct Losses Due to Floods and Droughts Figure 7. Mozambique: Long-term Direct Losses Due to Floods and Droughts Malawi: Direct Losses Mozambique: Direct Losses
PAGE 12
Results for Malawi indicate that, on average, droughts and oods together reduce total GDP by about 1.7 percent per year. Damages, however, vary considerably across weather events with total GDP declining by at least 9 percent during a severe 1-in-20 year drought.
PAGE 13
Measuring Economic Risks of Droughts and Floods | 11 Economy-wide Impacts of Droughts and Floods The study imposed estimated crop production losses on an economy-wide general equilibrium model of Malawi that captures farm and nonfarm sectors and households in different parts of the country. The model is based on information from a wide range of sources, including national accounts, foreign trade data, and a 2004-05 household survey. By integrating this information, the model estimated macro impacts on agricultural and national gross domestic product (GDP) and micro impacts on employment and household poverty. R esults for Malawi indicate that, on average, droughts and oods together reduce total G D P by about 1.7 percent per year. D amages, however, vary considerably across weather events with total G D P declining by at least 9 percent during a severe 1-in-20 year drought ( T able 1). Such severe outcomes place a signicant constraint on Malawis development prospects. Smaller-scale farmers in the southern regions of the country are especially vulnerable to declining agricultural revenues and rising poverty during drought and ood years (Box 2). U rban households also experience increased poverty due to higher food prices and declining nonfarm wages. Table 1. Malawi: Estimated changes in production (GDP) Droughts RP5 RP10 RP15 RP25 AAL Total GDP .5 .5 .2 .4 .0 Agriculture .1 .3 .9 .5 .0 Industry 0.0 0.0 0.3 0.7 0.0 Services .2 .3 .8 .4 .4 Floods* RP5 RP10 RP20 RP50 AAL Total GDP .7 .5 .2 .0 .7 Agriculture .5 .1 .5 .2 .4 Industry .6 .9 .2 .6 .2 Services .5 .7 .9 .2 .2 National results, although ood shock only applied to southern region. Poverty increases 0.9 percentage points during AAL oods Impact is more severe in the Southern region (2 percentage point increase) Percentage point increase poverty (from 52.4% in the base) RP5 RP25 AAL drought 0.7 16.9 1.3 (154,000 people) Declining household income and rising prices raise poverty Farm Non-farm NationalBox 2. Malawi: estimated changes in poverty rate 2.5 2.0 1.5 1.0 0.5 0.0%-pt Increase in Poverty Rate Urban Rural
PAGE 14
Model results estimate that droughts, on average, cause GDP losses of almost 1 percent every year. Economic losses are much higher during extreme droughts; for example, during a 1-in-25 year drought (RP25) similar to that experienced in 1991/92 in Malawi, GDP contracts by as much as 10.4 percent. Droughts also exacerbate Malawis already high levels of income poverty. On average, droughts cause a 1.3 percentage point increase in poverty, but this rises to almost 17 percentage points during an RP25 drought (this is equivalent to an additional 2.1 million people falling below the poverty line). Importantly when droughts do occur, their impacts vary considerably across regions and population groups with smaller-scale farmers most vulnerable to droughtinduced economic losses. Nonfarm and urban households are also vulnerable, especially the poor who spend a large proportion of their income on food. The impact of oods in the southern region of Malawi was also estimated. The average annual GDP loss due to oods is about 0.7 percent, thus making the average impact of oods slightly less than that of droughts. However, considering that this is the national-level impact of an event that is highly localized, i.e., one that only affects production levels in the southern region directly, the economy-wide effects are in reality quite severe. These national-level losses occur despite the fact that agricultural production in the central and northern regions may increase during oods. These benets arise from higher national food prices during southern oods. The implications for farming households in the southern region, however, are severe. Average annual crop and livestock losses range from 4 percent in Blantyre to 6.8 percent in Machinga. Floods are further found to mainly affect smalland medium-scale farmers.Concluding Remarks T he impacts of droughts and oods are often discounted or ignored in the long-tem national development planning and sectoral strategies development. Drought and oods, however, have considerable negative impacts on the economies of Malawi and Mozambique each year in terms of direct losses in assets, reduction of G D P, and poverty increase. T he results of the study also demonstrate that droughts and oods are major obstacles for agriculture and food security in both countries. Indications suggest that drought and ood events are becoming more frequent in Malawi and Mozambique, and thus the average annual impact might become even greater in the future. It is, therefore, crucial that policymakers take heed of the severe implications of climate variability, especially for the most vulnerable in society, such as resource-poor, small-scale farmers and poorer urban households. It is clear from this analysis that climate variability risks are important; they need to be considered and addressed explicitly in designing and evaluating national development policies and strategies in Malawi and Mozambique. T he assessments ndings and recommendations serve as a means to prepare disaster risk adaptation strategies or to expand existing national and sectoral policy and strategies. T he study has laid the groundwork for discussions and analysis of the effectiveness and viability of various measures to decrease economic vulnerability of the countries to the risks of droughts and oods. Preparation of detailed sectoral strategies to address the risks of droughts and oods will be undertaken through follow-up disaster risk management activities in both countries. 12 | Economic Vulnerability and Disaster Risk Assessment in Malawi and Mozambique
PAGE 16
Contact: Rimma Dankova rdankova@worldbank.org The World Bank 1818 H Street, NW Washington DC, 20433 USA T el: +1 (202) 473-1000 Fax: +1 (202) 477-6391 www.worldbank.org Contact: Murthy Bachu murthy.bachu@rmsi.com RMSI A-7, Sector 16 Noida 201301, INDIA T el: +91-120-251-1102, 2101 Fax: +91-120-251-1109, 0963 www.rmsi.com Contact: Karl Pauw K.Pauw@cgiar.org IFPRI 2033 K Street, NW Washington, DC 20006 USA T el: +1 202-862-5600 Fax: +1 202-467-4439 www.ifpri.org C O N TACT D ETAILS THE WORLD BANK
|
