URBAN FLOODING IN A CHANGING CLIMATE: CASE ...serialsjournals.com/serialjournalmanager/pdf/1436348337.pdfUrban Flooding in a Changing Climate: Case Study of Kolkata, India 137 to short,

Urban Flooding in a Changing Climate: Case Study of Kolkata, India 135

* Authors are respectively Lead Environmental Economist, Consultant and Senior Environmental Economistof the World Bank.

Asian-African Journal of Economics and Econometrics, Vol. 12, No. 1, 2012: 135-158

URBAN FLOODING IN A CHANGING CLIMATE:CASE STUDY OF KOLKATA, INDIA

Susmita Dasgupta*, Subhendu Roy* and Maria Sarraf*

ABSTRACT

Kolkata, a mega-city of India has been identified as one of the urban centers particularlyvulnerable to climate risks in the 21st century. Modest flooding during monsoon at high tide inthe Hooghly river is a recurring phenomenon in Kolkata. Climate change is likely to intensifythis problem through a combination of more intense rainfall, riverine flooding in the Hooghly,sea-level rise and coastal storm surges. This paper assesses potential damage to Kolkata fromincreasingly intense precipitation events of a 1 in 30/ 50/ 100 year return period and sea-levelrise for alternative scenarios of climate change by integrating information on climate changeand hydro-meteorological models with geographic overlays. Estimates indicate the annualexpected damage in Kolkata can be as high as US$ 5 billion, by 2050. Approximately 6.4 millionman days will be required to recover from the additional damage.

High resolution spatial analysis provides a roadmap for designing adaptation schemes tominimize impacts of such climate change. De-silting of trunk-sewers, construction of a stormwater retention infrastructure interlinking ponds and parks in the drainage infrastructure wouldcontribute to more efficient watershed management and minimize impacts from flooding. Beyondthese capital-intensive investments, improved policies, planning and institutions are needed toprotect vulnerable population in Kolkata.

1. INTRODUCTION

Projections by the Intergovernmental Panel on Climate Change and the World MeteorologicalOrganization suggest an increase in the frequencies and intensities of climate extremes in the21st century using various emission scenarios (WMO 2010; IPCC 2007). Heavily urbanizedmegacities in the low-lying deltas of Asia have been identified as “hotspots”, especially vulnerableto climate risks (ADB 2008; IPCC 2007). In many such cities, flooding during rainy season isalready a recurrent annual feature. Furthermore, poor inhabitants of these cities are among themost vulnerable as large and densely populated conglomerations of slums and shanties areinvariably located in areas of unplanned and unregulated development (World Bank 2010a,UNFCC, 2008). With the addition of the increased risk of storm surges, cyclones, and intenseprecipitation induced by climate change effects, such flooding conditions and associated impactcaused by weather related events may worsen dramatically into disasters (World Bank, 2010b).A recent collaborative study by the Asian Development Bank, the Japan International Cooperation

136 Susmita Dasgupta, Subhendu Roy and Maria Sarraf

Agency, and the World Bank on four coastal cities in Asia: Bangkok, Ho Chi Minh City, Kolkataand Manila has concluded that if current climate change continues, flooding is likely to occurmore frequently by 2050, submerging large parts of each city for longer periods of time; andcosts from major flooding events are estimated to run into billions of dollars (World Bank2010a).

In this paper, we present a study undertaken for the mega city Kolkata in India to assessthe impacts of climate change out to 2050. At present, the urban agglomeration, KolkataMetropolitan Area (KMA) as defined in the Vision 2025 document (KMPC 2004), featuresamong the 30 largest mega-cities of the world having population in excess of 10 million (UNDepartment of Economic and Social Affairs, 2005). Kolkata is vulnerable to a number ofnatural hazards including cyclones, tidal upsurge, urban storms, and intense local precipitationthat lead to water-logging and flooding. The scope of this paper is restricted to urban floodingmainly from increases in intense precipitation arising out of climate change as it is likely to bethe most significant climate change effect in an urban area like the Kolkata.

The remainder of the paper is organized as follows: Section 2 describes the study area,Kolkata. Modeling of impacts of intense precipitation in a changing climate to determine thetiming and magnitude of floods of different return periods along with vulnerable area andpopulation estimates are presented in Section 3. Section 4 documents potential damage fromflooding. A brief discussion of adaptation strategy is included in Section 5, and Section 6concludes the paper.

2. THE CITY OF KOLKATA

Kolkata (KMA), earlier known as Calcutta, is the capital of the State of West Bengal in India. Itcovers an area of 1,851 sq km with a population of 14.72 million as per the 2001 census.However, the study of Kolkata in this paper is limited to only the Kolkata Municipal Corporation(KMC), the more urbanized heart of the KMA due to paucity of available data for the entireKMA. With an area of 185 square Kilometers, the KMC is divided into 141 wards1 (KMC,2010). As per the 2001 Census, the population of the KMC is 4.6 million people with a populationdensity of 24,760 persons per sq. km. Existing land use pattern in the KMC is predominantlyurban reflecting 300 years of organic growth. In the KMC, residential or mixed residential areaaccounts for 68% of land use. With almost nonexistent land-use planning or control, residentialand non-residential uses co-mingle in most areas with little or no demarcations. Slums in theKMC are the hub of many informal manufacturing sectors some of which involve highly toxicindustries producing acids and other chemicals.

The KMC lies along the tidal reaches of the Hooghly and was once mostly a wetland area.Reflecting its previous character, a number of natural depressions remain; many of which aredead river channels. The elevation of KMC area ranges from 1.5 to 9.0 meters above the sea-level (masl) with an average elevation of 6 masl. While the KMA/ KMC are often perceived asa coastal city, in reality it is about 145 km away from the Bay of Bengal.

The KMC has a tropical wet-and-dry climate with an annual mean temperature of 26.8 °Cand monthly mean temperatures in the range of 19-30 °C. The region around KMC is subject


to short, high intensity precipitation, especially during the monsoon months between June andSeptember. This along with occasional coincidence of high tide is the usual cause of the urbanflooding. The annual rainfall is about 1,600 mm and the highest rainfall usually occurs duringthe monsoon, in the month of August.

Figure 1: Geographical Context of the Study

The KMC area is divided into nine major drainage basins, each with independent sewernetworks and a terminal pumping station. Three of the basins drain into the Hooghly river onthe west and 6 drain into Kulti system in the east. Eleven sluice gates on Hooghly prevent tidalingress during heavy storms and high tide into the sewer system. The existing sewer networkwith a length of 1,610 km and the length of open drain about 950 km covers 55% of its KMCarea. However, the central part of the KMC sewerage network system (Town System) is almost140 years old; heavy siltation and inadequate maintenance of the channel outfall structureshave resulted in a significant reduction in the hydraulic capacity of the KMC sewerage system(Kolkata Municipal Corporation, 2007).

The KMC, city of Calcutta (Kolkata), as described by Kipling, was “chance directed andchance erected.” The description is apt and relevant because of the challenges presented by itslocation and its exposure to flooding. At the outset, it should be mentioned that flooding inKolkata is an annual feature during the monsoons, whenever accompanied with major storms.There has been no mass exodus in Kolkata during temporary floods possibly because themagnitude of the floods has not been very large and the population has learned to adapt such as


taking care of critical assets and preparing for health risks as most of the area experiencesmoderate flooding on an annual basis. That situation may change if flooding becomes moresevere as a result of climate change impacts. The potential threats to the KMC from floodingcan be categorized as caused by

� Natural factors: The flat topography and low relief of the area that cause flooding inthe KMC area. The source of such flooding is high intensity rainfall, storm surge, andcyclonic storms.

� Developmental factors: These include unplanned and unregulated urbanization, lowcapacity drainage and sewerage infrastructure that have not kept pace with the growthof the city or demand for services, siltation in available channels, obstructions, mainlythrough uncontrolled construction in the natural flow of the storm water, reclamationof and construction in natural drainage areas (marshlands), etc.

� Climate change aspects: Changes such as increase in the intensities of rainfall, sealevel rise and the increase in the storm surges that may increase the intensity and durationof the flooding event.

3. MODELING CLIMATE CHANGE

The effect of climate change in the KMC area, as mentioned earlier, can reveal in a number ofdifferent ways. In the present analysis, only the hydrological and hydraulic impacts resultingfrom increased precipitation in a changing climate were studied as it is likely to be the mostsignificant climate change effect in an urban area like the KMC. In addition, storm surge andsea level rise from climate change were included as intense precipitation concurrent with hightide and extreme storm surge can cause increased flooding; and the result can be especiallydevastating for areas that are already vulnerable to flooding.

In order to model the impact of climate change on intense precipitation events, initially,three baseline scenario of flooding arising from historic precipitation level for 30 year, 50year, and 100 year occurrence2 were modeled assuming no climate change effects. The climatechange effects for 2050 were then added to these scenarios by multiplying the precipitation forthe 30 year, 50 year, and 100 year occurrence by a factor provided by the Japan InternationalCooperation Agency (JICA) for the A1F13 and the B14 emission scenarios respectively andusing the same precipitation distribution pattern as in the 30 year, 50 year, and 100 yearoccurrence without climate change effects. In order to take into account uncertainties inprojecting future climatic conditions, a high fossil fuel intensive (A1F1) and a low emission(B1) scenario were considered as a starting point of this analysis. An estimate of expected sealevel rise of 27cm by 2050 was also included in the climate change scenarios(World Bank,2011). In all these scenarios with and without climate change effects, the study then assessedthe impact in terms of the extent, magnitude, and duration of flooding.

Models

In assessing the magnitude of flooding events, the key factors considered were the inflow andoutflow of water in and around the KMC area. The inflow depends on the precipitation in theKMC area, the overtopping of the Hooghly river due to water inflow from local precipitation as


well as that from the catchment area, and storm surge effects. For the outflow, importantconsiderations are the natural discharge through drainage basins and sewerage systems in placeas well as installed pumping capacity. The rate of discharge is also affected by tide levels andstorm surge effects. An imbalance between inflow and outflow, especially caused by shortduration intense precipitation, results in local flooding as the water inflow overwhelms thenormal drainage, sewerage, and pumping capacity.

Three models: hydrological, hydraulic and Urban Storm models were used to capture theeffect of all factors that lead to flooding in the KMC. A hydrological model: Soil and WaterAssessment Tool (SWAT) was used to estimate the water flow in the Hooghly river system.The water flow was modeled using the daily gridded rainfall and temperature data obtainedfrom India Meteorological Department for the whole Hooghly River catchment for a 25 yearperiod and water flow from diversions made into Hooghly from Ganga River 257 kms upstreamfrom Kolkata. The generated data, daily flow series at various locations along the HooghlyRiver was then fed into a hydraulic model: Hydrologic Engineering Centre – River AnalysisSystem (HEC-RAS) to generate flood waves moving through the river channel (U.S ArmyCorp of Engineers, 2002).. The tidal and storm surge effects were fed into the HEC-RASmodel as boundary conditions. Output from the model provided the water surface profiles allalong the river coupled with change in flow depth during the flood period. Finally, a dynamicrainfall-runoff simulation urban storm model: Storm Water Management Model (SWMM)was used to simulate the flooding that will result once the river flooding is combined withlocal rainfall and drainage capability of the KMC area, by incorporating the prevailing urbancharacteristics such as the sewerage in place, pumping capacity of drainage pumps, and thelikely operation of lock gates during the flooding period etc.

Modeling Scenarios

In the baseline (without climate change) scenario, the return periods of precipitation events inKMC was based on available historical rainfall data for 25 years. The data obtained fromcontinuous recording rain gauges, at each successive 15-minute interval, for the monsoon period(April to September) from 1976 to 2001 were processed to extract maximum rainfall eventscorresponding to different storm durations for the two recording stations.

Inputs for the climate change scenarios for Kolkata, provided by Japan (JICA, 2008) werebased on pattern scaling techniques applied to 16 Atmospheric-Ocean General Circulation Modelsused for IPCC Fourth Assessment Report (for details, see Sugiyama, 2008)). The scenariopredictions included a temperature increase in Kolkata of about 1.8 °C for A1F1 scenario and 1.2°C for B1 scenario. Precipitation predictions were provided as a fractional increase in theprecipitation extremes of about 16 per cent for A1F1 and 11 per cent for B1 scenario imposedabove the baseline distribution of precipitation. A sea level rise of 27 cm by 2050 was also addedto the storm surge for the A1F1 and B1 climate scenarios (World Bank, 2011).

Model Calibration and Validation

In order to calibrate and validate the models, the theoretical distribution of rainfall patterns fora 100 year return period from Log Pearson Type III distribution was compared with observed


15 minute rainfall intensities in the study area from 1976 to 2001. A visual comparison ofhyetographs with the 1978 rainfall pattern on September 27, 1978 from 5:03 PM to 8:00 PM, aperiod with the highest rainfall intensity in recent times, showed a good match with the theoretical100 year return period rainfall.5 It is thus found that the return period of precipitation event of1978 that caused the most damaging flood in Kolkata in recent times can be used for the 100year return period precipitation.

The flow obtained from the SWAT model for flood peaks was also matched withtheoretically generated flow from a number of distributions. Of these, the distribution thatfitted best was the Gumbel Extreme Value Type I distribution. With a resultant R2 of 0.98between the two variables, the simulated flow using SWAT and the synthetic flow using Gumbeldistribution, the match was found to be near perfect. The peak flow for all other return periodswere therefore based on the Gumbel distribution.

The analysis of the flood events requires the complete rainfall distribution (rather than justthe peak used to fit the theoretical distribution). To remedy this, the complete flood eventswere obtained by applying a ratio of rainfall corresponding to the respective return periodsderived using the rainfall analysis. The observed rainfall time series was then updated withthese ratios and SWAT simulations were carried out for each of these return periods.

Model Simulation Results

The models were set up and run for the baselines of 30 year, 50 year and 100 year return periodfloods without climate change as well as for the climate change scenarios A1F1 and B1(superimposed on a 30 year, 50 year, and 100 year return period storm).

Percentage of the KMC area inundated under different flood depth categories, as shown inTable 1, clearly indicates the largest affected area for the baseline scenarios belong to the twodepth categories: 0.25 – 0.50 m and 0.50 – 0.75 m, without climate change. However whenclimate change effects are added for scenarios A1F1 and B1, the flooding depth goes up byone notch with the depth categories 0.50 – 0.75 m and 0.75 – 1.00 m, accounting for the largestaffected areas. This increase in the depth of flooding in the KMC is mainly due to its topographythat prevents natural drainage.

Since flood water depths below 0.25 m cause little damage to the affected area andpopulation, we have aggregated the percentage of area and population exposed in the KMCwith flood depths of 0.25m and more under various scenarios, and presented the estimates inTable 2. In the KMC, 39% of the population is exposed to flood water depths above 0.25 m inthe 30 year return period and 45% in the 100 year return period. With climate change, thepopulation affected increases to 41% under both 30yr+B1 and 30yr+A1F1 scenarios. Under100yr+B1 and 100yr+A1F1 scenarios the percentage goes up to 47%. See Table 2.

Figure 3 provides a comparison of the duration of inundation of various depths betweenthe 100 year return period without climate change effects and the A1F1 climate change scenarioadded. It shows that flooding persists even after 10 days in large parts of the KMC. In addition,the difference in extent and depth of flooding between the 100 year return period and the A1F1added scenario continues even after 10 days (see Figure 2).

Urban Flooding in a Changing Climate: Case Study of Kolkata, India 141T

able

1P

erce

ntag

e of

Are

a In

unda

ted

in t

he K

MC

Flo

od d

epth

ran

ge (m

)30

yr

30yr

+A

1F1

30yr

+B

150

yr

50yr

+A

1F1

50yr

+B

110

0 yr

100y

r+A

1F1

100y

r+B

1

0.00

- 0.

101.

00.

30.

30.

70.

30.

30.

30.

20.

20.

10 -

0.25

4.7

1.7

1.7

3.9

1.5

1.4

2.8

1.9

2.2

0.25

- 0.

5013

.67.

38.

511

.36.

46.

910

.47.

68.

10.

50 -

0.75

9.3

12.1

9.9

11.9

11.9

12.2

15.5

13.9

15.1

0.75

- 1.

002.

76.

35.

14.

08.

27.

47.

211

.110

.01.

00 -

1.50

2.2

3.5

3.6

2.5

4.4

3.9

4.1

5.9

5.3

1.50

- 3.

001.

11.

51.

31.

41.

81.

72.

02.

22.

1>

3.0

0.0

0.1

0.1

0.0

0.3

0.2

0.2

0.4

0.3


Table 2Exposed Area and Population to Flood Depths Higher than 0.25 m

Flood return period scenarios

30yr 30yr+A 30yr+B 50yr 50yr+A 50yr+B 100yr 100yr+ 100yr+1F1 1 1F1 1 A1F1 B1

KMC: Area 34% 35% 34% 36% 39% 38% 39% 41% 41%affected (%)

KMC: Population 39% 41% 41% 42% 43% 44% 45 % 47% 47%affected (%)

Figure 2: Duration of inundation of various depths in all of KMC for 100 Year and A1F1 Scenario

It should be noted that land subsidence in the KMC was not included in the present analysisbecause of inherent uncertainties about future changes. For the KMC, the literature indicatesthat some areas had been undergoing subsidence ranging from 6.52 - 13.0 mm per year onaverage for a period of 42 years from 1958 to 2000 (Nandy 2007; Chatterjee et al., 2006). Ifsuch subsidence continues over a longer period of time, land subsidence data input will become


important in modeling as it may drastically increase extent of flooding. On land subsidence,this present study concludes with a cautionary note that land subsidence can further exacerbatethe flooding caused by climate change effects.6

4. DAMAGE ASSESSMENT

Flooding in an urban area like Kolkata can lead to widespread damage resulting from submersionof building and property as well as disruption of livelihood among the flood affected population.The damage is further enhanced the longer the flooding persists. While impacts like damage toproperty and loss of income from flooding can be more readily quantified, it is often difficult toassign monetary values to many other types of damages that lead to the dislocation of normalliving through socio-economic, environmental, and health impacts. An effort has been made inthis paper to account for readily quantifiable monetary damages as well as to impute a monetaryvalue wherever possible to other intangible damages (ECLAC 2003). The analysis focuses onthe likely impact of flooding arising out of climate change effects on the affected population inthe KMC in the year 2050. The selection of the time horizon 2050 is appropriate given citylevel planning horizons in India and the typical time frame for major flood protection measures,which is about 30 years.

The analysis begins with an estimate of damage from 30 year, 50 year, and 100 year floodoccurring in 2050 in current climate. The study then examines the additional damage that canarise if the effects of climate change as envisaged in the A1F1 scenario are added to that 30year, 50 year, and 100 year flood event respectively. In the analysis, all relevant data is projectedfor the KMC in the year 2050. For example, the expected population size of the KMC in 2050is extrapolated based on the past decadal growth rates adjusted for likely future changes(JNNURM 2005). The valuation of property and income levels in 2050 are also estimatedusing the average per capita GDP growth rates in the recent past and projections for the future.To assess the damage in real terms, all data are based on 2009 prices and thus ignores anyinflation that can occur by 2050.7

In general, both the depth and duration of a flood play a role in the damage it causes. Thedepth of an inundation has more effect on the physical capital; and if the depth exceeds acritical level then duration of a flooding has greater impact on the day to day livelihood of thepopulation. Since the severity of damages from flooding is critically linked to the depth ofwater inundation, four water levels were identified and associated damage severity was derived.As mentioned earlier, it has been found that a depth level below 0.25 m produces little damagein most affected areas as people in the KMC have learned to adapt to such level of flooding asa common occurrence every year. However, when the depth of flooding exceeds 0.25 m, somedamages are unavoidable. In general, it has been observed that the extent of damage to assetsis mainly dependent upon the depth of flooding with: 0.25 – 0.75 m leading to a low impact;0.75 – 1.50 m leading to a moderate impact, and > 1.50 m causing an extensive impact. Suchdamage levels also continue to rise the longer the flooding persists.

Since the extent of flood damage also depends on the duration of flooding, the duration offlooding above 0.25 m level is analyzed up to 10 days in order to examine the severity of flooddamage. The cut-off of 10 days was chosen as damage caused by flood water generally reaches


the upper threshold of damage caused if flooding persists that long. Moreover, it was foundthat water starts receding rapidly from most areas in the KMC by that period even for the A1F1scenario with the 100 year flood.

Damage Assessment Methodology

The damage is estimated separately for the following main sectors of the economy: residential,commerce and industry, health, road, transport, electricity, telecommunication, and water-supply.In each sector, efforts have been made to estimate the damage to physical capital and the damageto earnings to get an estimate of total damage.

The damage to physical capital is assessed by first computing the total value of thephysical capital, then determining the percentage of physical capital likely to be exposed toflooding, and finally applying a damage factor that is based on the depth and duration offlooding.

Damage to Physical Capital = Value of Physical Capital* Percentage ofPhysical capital Exposed to Flooding*Damage Factor (4.1)

The damage to earnings is assessed by first determining the daily value of income/outputgeneration and multiplying that by the number of work days lost due to flooding.

Damage to Earning = Daily Value of Income/Output*Average Number ofDays of Loss (4.2)

Residential Sector

Computation of the residential damage is based on the damage to physical capital (buildingsand property) and income losses.

Building Damage: Damages to residential buildings due to flooding have been separatedinto (i) repair costs and (ii) clean-up costs.

(i) Repair costs: The repair costs are estimated based on replacement costs of the damagedbuildings. The proportion of building exposed to damage will depend on the depth andduration of flooding as well as the nature of the building. To estimate the value ofresidential buildings of the KMC, Google Earth data is used to separate the compositionof the city residential buildings into the three broad categories of housing: EconomicallyWeaker Sections (EWS), Middle Income Groups (MIG) and High Income Groups(HIG). The composition for 2050, EWS: 22%, MIG: 57% and HIG: 21% is then assessedbased on a percentage estimate of EWS moving out of core city areas. Discussion withthe town planning authority in the KMC suggested that the average floor areas forEWS, MIG and HIG are 25 sq m, 75 sq m, 150 sq m respectively. The buildingconstruction costs in Kolkata per sq m of floor area are found to be $156 per sq m forEWS, $ 181per sq m for MIG, and $ 249 per sq m for HIG based on the ACC HelpHome Building Calculator at 2009 prices (ACC 2009). Using the floor area for eachcategory of residential building, the construction costs for EWS, MIG and HIG buildingsat 2009 prices are calculated. The building construction costs are then extrapolated to2050 costs using an annual growth rate in construction costs.


Based on past experience, it is expected that residential building damage from floodingwill occur only at the first floor and not at the upper floors in a multi-storied building.Observation reveals that on average, EWS are two storied, MIG are three storied, andthe HIG are four storied buildings. Any inundation of depth > 0.25 meter for more thana day is expected to cause a minimum threshold proportional damage requiring repairthat varies with the type of the building (see Table 6). The minimum damage thresholddiffers across EWS, MIG and HIG buildings due to the quality of building materialused in the construction. It is also assumed that any building submerged for 10 days ormore (with inundation depth > 1.5 m) will always require repair. The proportion ofbuildings requiring repair decreases linearly for fewer days of inundation. For depthsof 0.75 m – 1.5 m and 0.25 m – 0.75 m, the proportion of damage is assumed to be 75%and 50% respectively of the proportion for depth > 1.5 m. Since the depth of inundationkeeps changing as water recedes from a given area, the proportion of building requiringrepair in an area is the maximum proportion found from the inundation duration for thethree different depths over a 10 day period.

(ii) Clean-up costs: It is assumed that all inundated buildings not requiring repair willincur clean-up costs. Clean-up costs mainly involve disinfecting building premisesand minor repairs including cleaning of the floor, whitewashing walls etc. Based onfield surveys, Clean-up costs are assessed to be around $ 12, $ 45, and $ 90 respectivelyfor EWS, MIG, and HIG. These were then extrapolated to 2050 costs.

The total damage for residential buildings is then determined by combining the buildingrepair and clean-up costs. For details, see Table 3.

Table 3Data for Residential Buildings

Type Average Average % of Proportion Building Repair costs Minimum Clean upfloor area no. of story building in of Building cost ($’000) per Damage costs

(m2) per building each floor area ($Million) building in Threshold ($’000)category in first 2050 per

floor11 buildingin 2050

EWS 25 2 22% 0.5 0.019 1.11 33% 0.056MIG 75 3 57% 0.33 0.068 4 25% 0.22HIG 150 4 21% 0.25 0.186 11.11 20% 0.44

Property Damage: Experience suggests that extensive residential property damage is causedby flooding, especially if it is sudden and prolonged. All household property like vehicles,appliances, electronic goods, furniture, and other belongings can be damaged to varying degrees,especially if they cannot be moved to upper floors. Since there is a direct correlation betweenhousehold income and the value of household property owned, the estimate of residentialproperty is made based on the household income in the KMC. In particular, it is assumed thatthe value of the property owned depends on the savings accumulated in each income categoryover a 5 year period. These are then extrapolated using an average annual growth in income to


arrive at 2050 income levels (see Table 4). The midpoint in each income range is used as theaverage income in that range other than the first and the last range because of the skeweddistribution in those two ranges.

Table 4Household Data in KMC

Income category ($)12

< 1700 1700-150 150-300 300-500 500-1000 >1000

Percentage of Household 7.9 25.2 26.4 27.4 8.0 5.0Average income in 2050 ($’000) 5.6 12.4 24.9 44.4 83.1 222.0Savings rate 5% 7.5% 10% 15% 20% 25%Proportion of household in first floor 0.5 0.33 0.33 0.25 0.25 0.25

The damage to residential property is estimated based on the total property exposed toinundation in the first floor times a property damage factor. The proportion of households thatface property damage in a Ward is given by the maximum percentage of area flooded in thatWard. The property damage factor, on the other hand, is based on the depth and duration offlooding. Based on observation, it is assumed that a maximum of 33% of household propertywill be damaged if buildings are submerged for 10 or more days with inundation depth > 1.5 mwith proportional decrease for fewer days of inundation. The corresponding maximum damagefor depths 0.75 m – 1.5 m and 0.25 m – 0.75 m are assumed to be 25% and 20% respectively.The final damage factor uses the maximum damage factor found from the inundation durationfor the three different depths.

Income Loss: Flooding in the KMC affects the income of both the KMC residents anddaily migrant workers who commute to the KMC for their livelihood. While the population ofthe KMC is estimated at 4.6 million as per the 2001 census, another 6 million people areestimated to commute to the KMC daily for work. For the purpose of our analysis, the populationof the KMC as well as the number of migrant workers is extrapolated to 2050 using estimateddecadal population growth.

The estimates of income loss exclude any income loss in the organized sector because oftwo reasons. Firstly, most workers in the organized sector are paid on a monthly basis (such asoffice workers and those working in the administration) and hence do not face loss in incomefrom daily disruptions. Secondly, workers in the organized sector, who are paid on a dailybasis, are likely to face income loss, but this loss is already captured under the commerce andindustry sectors. It is therefore not included here to avoid any double counting.

Household income data is used to estimate income loss for KMC residents working in theunorganized sector, such as artisans or construction laborers. The income loss is based on thenumber of lost work days (average duration of flooding in all of KMC) multiplied by theaverage income in each category. In the absence of secondary data on workforce in unorganizedsectors, for the purpose of this analysis it is assumed that 90% of households in the lowestannual household income bracket (<$1,700), 50% of households in the medium householdincome bracket ($ 1,700- $3,400 ) and 10% of households in the higher household income


bracket ($ 3,400 - $6,800 ) are employed in the unorganized sector. It is also assumed that 25%of migrant workers commuting daily to the KMC are in the unorganized sector. Since workersin the unorganized sector tend to be less skilled, it is further assumed that migrant workers inthe unorganized sector earn 33% less on average than the average urban worker residing in theKMC. The income loss for migrant workers is computed based on the number of lost days ofwork (which is based on the average duration of flooding in all of KMC).

Total Damage in Residential Sector

Faced with a 100 year flood in 2050, an estimated 14.2% of all household in KMC will facevarying degrees of damages to their residential building (0.16 million out of total 1.13 millionhouseholds) even without climate change. Adding the impact of climate change under an A1F1scenario will increase this number to 15.2% of all households (or 0.17 million). Damages toresidential buildings are estimated at $ 550million in the current climate and at $ 620million underthe added A1F1 scenario with a 100 Year return period event, at $ 490 million in the currentclimate and at $ 550 million under the added A1F1 scenario with a 50 Year return period event, at$ 470 million in the current climate and at $ 530 million under the added A1F1 scenario with a100 Year return period event. In terms of household property losses, damages are estimated at $760 million (for current climate) and at $ 860 million (under the added A1F1 scenario) with a 100Year return period event, at $ 680 million (for current climate) and at $ 770 million (under theadded A1F1 scenario) with a 50 Year return period event, at $ 650 million (for current climate)and at $ 740 million (under the added A1F1 scenario) with a 30 Year return period event.

The total loss of residential income from the 100 year flood in the current climate is estimatedat $ 100 million, of which the KMC residents will suffer a loss of $ 30 million while migrantworkers will lose an income of $ 70 million. Under the added A1F1 scenario, the total incomeloss increases to $ 110million and the major share of that loss is faced by migrant workersamounting to $ 80million. Under the 50 Year flood the total income loss is $ 90 million (forcurrent climate) and at $ 100 million (under the added A1F1 scenario) and under the 30 Yearflood the total income loss is $ 80 million (for current climate) and at $ 90 million (under theadded A1F1 scenario).

Commerce and Industry Sector

The damage to physical capital in the commercial sector is based on the damage to buildingsand merchandise, and in the industrial sector on the damage to buildings and machinery asproxied by fixed capital. The average cost of a building in a commercial establishment isdetermined by using an average area of 50 sq m and a construction cost of $ 160 per sq m. as perthe ACC Building Calculator. In the commercial sector, the merchandise in stock at any pointof time is estimated based on holding necessary to meet a week’s sale. Assuming that thecommercial buildings on average are two storied, since only commercial establishments locatedin the first floor will face flood damage, it is assumed that only 50% of the building andmerchandise will face flood damage. For industries, the fixed capital data is used as an estimateof the value of building and machinery. All damage costs are then extrapolated to 2050 costsusing an annual growth rate.


In the absence of ward wise distribution of commercial and industrial data, the damagefactors for commerce and industry is developed using the average extent of flooding for 10days under various depths for the whole of the KMC area. The damage to commercial andindustrial buildings is then determined by applying the building damage factor and the damageto merchandise and machinery by applying the property damage factor developed in the sameway as in the residential sector.

The loss of earnings in commerce and industry is assessed as the difference in the daily netvalue added and the avoided variable costs for the number of days the business is affected. Thenumber of lost days of business is based on the extent of average flooding and its duration inthe KMC.

Total Damage in Commerce and Industry Sector

Estimates indicate that faced with a 100 year flood in 2050 in the current climate, expecteddamages to commercial sector and industrial sector are $ 170 million and $ 60 millionrespectively. Adding the impacts of climate change under the A1F1 scenario, such damages areexpected to increase to $ 210 million and $ 70 million respectively. With a 50 year flood thedamages to commercial sector and industrial sector are $ 140 million and $ 50 million respectively(for current climate) and $ 190 million and $ 60 million respectively (under added A1F1 scenario)and with a 30 year flood the damages to commercial sector and industrial sector are Rs $ 130million and $ 40 million respectively (for current climate) and $ 190 million and $ 60 millionrespectively (under added A1F1 scenario).

Health Sector

Flooding causes damages to hospital infrastructure and increases the incidence of water bornediseases. Due to limited data availability, the analysis could not include damage to healthinfrastructure. Increased incidences of water borne diseases are evaluated by assessing both theadditional treatment costs as well as the loss of productivity due to illness.

Treatment costs: Annual incidence of major water borne diseases in the KMC totals around5.86% (Gastroenteritis, Cholera: 3.46%, Hepatitis: 0.05%, Typhoid: 0.18%, Malaria,Chickengunya Dengue: 2.17%).8 While anecdotal evidence is available about the higherincidence of such disease due to increased amount of stagnant water from flooding, actual datato estimate the impacts of flooding is hard to come by. However, during the rainy season andits immediate aftermath it has been observed that incidences of water borne diseases more thandouble. Hence, for the purpose of this analysis, a conservative estimate of doubling of theincidence of water borne disease (to 11%) in the flooded area is used for both the 100 Yearflood and the A1F1 scenario. Average per capita health care costs for treating such waterborne diseases are estimated around $ 10 at 2009 prices.

Mortality and morbidity that may occur due to extreme heat events as a result of higheraverage temperatures arising from climate change effects has not been included in the analysis.

Increased morbidity: The increased incidence of diseases also entails a loss of productivity.Various illnesses are converted into one common indicator, the Disability Adjusted Life Years


(DALY), for ease of computation. In West Bengal, an estimated 171 DALYs per thousandurban population is estimated to be lost every year due to various diseases (KEIP, 2007).Current data indicates that more than 7% of such DALYs can be traced back to water bornecauses like diahhroea and malaria (WHO 2004). These data are projected to 2050 and it isestimated that around 14 DALYs per thousand people will be lost due to water borne diseasesin the KMC in 2050. In order to factor in the impact of climate change, a doubling of incidenceof such diseases in the flooded area is used for both the 100 Year flood and the added A1F1scenario. In order to value economic losses due to morbidity, a value of a statistical life of $0.33 million computed through a survey among workers in Chennai and Mumbai byS. Madheswaran, 2007, is used.

Total Damage in Health Sector

As explained above, a flood usually leads to an increase in the incidence of water bornediseases generating additional treatment costs and a loss in productivity. With no climatechange effects, a 100 year flood in 2050 will result in $ 13 million treatment costs and a lossin productivity amounting to $ 370 million. In a changing climate, under A1F1 scenario thetreatment cost is expected to increase to $ 16 million and the loss of productivity will be $390 million. The total effect arising from treatment cost and productivity loss under a 50 yearflood amounts to $ 340 million respectively (for current climate) and $ 380 million respectively(under added A1F1 scenario). Under a 30 year flood the corresponding damages are $ 330million respectively (for current climate) and $ 370 million respectively (under added A1F1scenario).

Road Sector

Floods cause damage to the road network (physical capital) and disrupt economic activity dueto delays, congestion and road closure resulting in loss of revenue. Due to data constraint, onlythe damage to road infrastructure has been estimated in this analysis based on repair costs.Experience suggests that the extent of damage to roads depends primarily on the depth offlooding over the road. In order to determine the road repair costs, Kok’s estimates of damagefactors of 0.1, 0.225, and 0.4 for water depths of 0.25-0.75 m, 0.75-1.5m, and > 1.5 m respectivelyare used (Kok, 2001). In the absence of location specific inundation data, the damage to roadsis determined based on the average depth and extent of flooding in the KMC applied to the totalroad length of 1,650 km in the KMC. The maximum extent of inundation in the KMC area over10 days of flooding under each depth of flood is then used to determine the extent of damagedroads.

Average road repair and rehabilitation costs of $ 0.2 million per kilometer, anestimate based on the World Bank transport data used for projects in India (WorldBank 2010c), is used in the computation. This estimate is further validated using actualdata from “Project-Special Road Repairing Works” of the KMC9. The average roadrepair costs are extrapolated to 2050 for the final analysis. The road repair cost caused byflooding is then obtained by multiplying the length of damaged roads by the per km roadrepair cost.


Total Damage in Roads Sector

Of the total road length of 1,850 Km in KMC (KMC 2010), a 100 year flood will cause varyingdegree of damage to 579 Km of roads and the length of road damaged will increase to 694 Kmunder the added A1F1 scenario. The total road repair cost amounts to $ 45 million for a 100year flood in 2050 with no climate change effects. This amount increases to $ 55 million underthe added A1F1 scenario. Under the 50 year flood, road repair costs amount to $ 42 millionrespectively (for current climate) and $ 50 million respectively (under added A1F1 scenario)and under the 30 year flood, road repair costs amount to $ 40 million respectively (for currentclimate) and $ 48 million respectively (under added A1F1 scenario).

Transport Sector

The damage of physical capital in the transport sector arises primarily from the damage to vehicles.In computing damage to vehicles, all powered public transport vehicles including buses, taxis,auto-rickshaws, and goods vehicles are included. Due to data constraint, damage to the railwaynetwork as well as to the airline infrastructure and disruption were not includes in the analysis.Although these impacts are not expected to be major, it should be noted that omission of thesefrom the analysis is likely to underestimate the overall impact for a flood event.

The number of public transport vehicles in 2050 is estimated using the estimated decadalgrowth rate of population in the KMC. To assess the repair cost for damaged vehicles due toflooding, data is obtained using sample surveys from auto repair shops. The average repaircost estimates range between $ 225 – 670 based on the depth of water submersion and the typeof the vehicle. To determine the proportion of vehicle damaged by flood, it is assumed that50% of the vehicles in depth category > 1.5 m, 33% in the depth category 0.75 m - 1.5 m, and25% in the depth category 0.25 m - 0.75 m will suffer damage.

All revenue loss estimates in the road transport sector are made using data for annual trafficand revenue for buses and trams owned by the Calcutta State Transport Corporation (CSTC) andthe Calcutta Tramways Corporation (CTC) respectively. The loss of income is calculated basedon the loss in daily revenue offset by the savings in variable costs. It is further assumed that therecan be no transport in areas where the depth of flooding exceeds 1.5m, with the correspondinglosses for the depths of 0.75 m – 1.5m and 0.25 m – 0.75 m being 67% and 50% respectively.

The net earning loss is computed based on the passenger-km traveled and the tariff chargedby the various modes of transport. The passenger-km travel data is available for buses; and fortaxis, auto-rickshaws, and goods vehicles - the data for buses is used with proportional reductionfor passenger carrying capacity. The revenue loss for goods vehicles is computed using halfthe distance traveled by buses along with per km rates for transport. The losses to intercitytrains are calculated using the average daily revenue loss derived from the annual revenue.The loss of revenue for Kolkata Port is also calculated using the average daily revenue and thenumber of days lost because of disruptions caused by flooding.

Total Damage in Transport Sector

Transport sector is facing an estimated damage to vehicles amounting to $ 40 million and anincome loss of $ 35 million, when faced with a 100 year flood in 2050 under current climate.


Expected vehicle damages and income losses will increase to $ 43 million and $ 43 millionrespectively under added A1F1 scenario. The total damage under a 50 year flood amounts to $62 million respectively (for current climate) and $ 78 million respectively (under added A1F1scenario). Under a 30 year flood the corresponding damages are $ 58 million respectively (forcurrent climate) and $ 75 million respectively (under added A1F1 scenario).

Electricity Sector

Past experience reveals that coal based power generating units face disruption due to floodingof coal handling plants and outages arising from flooding of machinery. While majority ofcurrent power plants in Kolkata is currently coal based, it is difficult to predict the likelyscenario in 2050, especially because of the concern about global warming effects from coalbased power plants. Moreover, in Kolkata due to their proximity to water bodies needed forcooling, power generating units take extra precaution for flooding. So, damage to physicalcapital of power plants from flooding is not expected to be large, and hence excluded fromthis analysis.

The electric utilities in Kolkata are still susceptible to loss of revenue from the need todisconnect electric supply in severely submerged areas to prevent accidental electrocution.The loss of income from such effects is calculated using the revenue loss in the flooded areaswhich is offset by a factor to account for the savings in variable input costs.

Total Damage in Electricity Sector

Under the 100 Year flood in current climate, the estimated total income loss of electricity sectoramounts to $ 7 million in 2050 and it increases to $ 8 million under the A1F1 scenario. Thedamage under a 50 year flood amounts to $ 6 million respectively (for current climate) and $ 8million respectively (under added A1F1 scenario). Under a 30 year flood the correspondingdamages are $ 6 million respectively (for current climate) and $ 8 million respectively (underadded A1F1 scenario).

Telecommunication, Water Supply and Other Public Sector Damage

In the telecommunication sector, flooding affects the landlines more than wireless. Since landlinesare currently witnessing a decline being substituted by wireless, it is expected that mosttelecommunication service will be provided through wireless by 2050. So, the effect of floodingon telecommunication is assumed to be minimal.

Any damage to water supply and sewerage infrastructure from flooding is also likely to beminor and difficult to assess. Since the water rates in KMC are based on the ferrule size of thewater line and not metered, any disruption of water supply caused by flooding will not lead toany significant revenue loss.

Increased flooding will affect government buildings and other government ownedproperty. It will have an impact on school and hospital buildings. both private and governmentowned. Such damages, although substantial, are difficult to assess and are not included inthe study.


Expected Damage Estimates in 2050

The sector-wise break down of total damages under the 100 year, 50 year, and 30 year returnperiod flood is shown in Table 5. To compute the expected damage in 2050 based on theseresults, these damages with the probability of each of those flood events are combined. Theprobability for a return period of x year flood occurrence is 1/x. So, if y is the total damage in2050 for a flood occurrence with x year return period, then the expected damage E in 2050 fromall such events up to a maximum flood return period of 100 is given by

100 ( 1)

1

1ln100

a bE e x dx�� (4.3)

where to estimate the damage y for each flood return period of x year we use a logarithmicregression

In y = a + b lnx (4.4)

Table 5Total Losses in Major Sectors in KMC (in 2050) in $ Million

Rs M 30 Year 30 Year 50 Year 50 Year 100 Year 100 YearFlood A1F1 Flood A1F1 Flood A1F1

Scenario Scenario Scenario

Residential Building 470 530 490 550 550 620Residential Property 650 740 680 770 760 860Residential Income Loss 80 90 90 100 100 110Commerce 130 185 140 190 170 210Industry 40 55 50 60 60 70Health Care 330 370 340 380 383 406Roads 40 48 42 50 45 55Transport 58 75 62 78 75 86Electricity 6 8 6 8 7 8Total 1,804 2,101 1,940 2,186 2,150 2,425

Overall Damage

The expected total damage in 2050 taking into account flood events up to a 100 year returnperiod along with their likelihood of occurrence is found to be $ 1.56 billion and it increases to$ 1.72 billion under the A1F1 scenario. The additional damages resulting from the added A1F1climate change scenario in 2050 is thus estimated to be $ 160 million, approximately 1.1% ofGDP of KMC in 2050 (ReDiff Business 2010). For a meaningful comparison, it is appropriateto convert the loss in local currency to US $ using the Purchasing Power Parity index (PPP).Because of the uncertainty about the likely PPP index in 2050 and as the losses for 2050 areexpressed in 2009 prices, the PPP adjustment is made based on the PPP index for India in 2009of 2.88 (IMF 2009). Using PPP US$, the expected damage under current climate amounts to$4.39 billion and with the A1F1 scenario it increase to $4.86 billion (see Figure 3). So, theadditional loss from climate change effects under the A1F1 scenario in KMC area in 2050amounts to $470 million.


Another useful way of assessing the extent of damage from climate change effects is toestimate the number of man days necessary to recover from the loss. This computation isbased on the average daily wage of urban workers in Kolkata (JNNURM 2009). For thiscomputation, the data for 2009 is projected to 2050 income levels using the per capita growthin GDP in the recent past. It is found that due to climate change effects under the added A1F1scenario in the KMC area in 2050, the additional damage in terms of man days will amount to6.4 million (see Figure 3).

It should be noted that the estimated additional damage of $470 million in KMC fromclimate change effects, presented in this study, is based only on the damage resulting fromincreased flooding and leaves out impacts from other weather related incidents like cyclones.Land subsidence is also not included in the analysis as it is considered a second order effect inconnection with the increased damage from climate change. Many other impacts could not bequantified in this analysis also due to data constraints. In addition, the damage estimates arebased on a partial equilibrium analysis and do not include losses in consumer surplus. Hence,it is important to note that this estimate is likely to underestimate total damage due to climatechange and the estimated damage of $470 million therefore represents a lower bound andactual damage is likely to be even higher.

5. ADAPTATION STRATEGY

Modest flooding during monsoon at high tide in the Hooghly river is a recurring hazard inKolkata. The local population has learned to adapt by developing a number of coping strategiesfor facing such periodic episodes of flooding. However, as discussed in section 3, more intenserainfall, riverine flooding, sea-level rise, and coastal storm surges in a changing climate can

Figure 3: Total expected damages in 2050 using PPP and Man Days


lead to widespread and severe flooding and bring the city to a standstill for several days.Vulnerable wards (reference earlier paper).

Undeniably, a major cause of periodic flooding of Kolkata during the rainy season is the‘Adaptation Deficit’ that Kolkata faces at present to cope with intense precipitation events.Major adaptation deficits in Kolkata currently include deficits in (a) sewerage network andtreatment infrastructure, (b) drainage infrastructure, and (c) financial resources and institutionalcapacity. The capacity of the sewerage systems in the KMC have not kept pace with the changesin population as the city has evolved. Almost all major pumping stations operate at much lessthan their rated capacity due to inadequate maintenance and renovation of equipment andbuildings; and the hydraulic capacity of the outfall canal system has been reduced due tosiltation and deposition of solid waste.

Kolkata needs a comprehensive and effective adaptation strategy with investment in both“soft” (institutions and policies) and “hard” (capital-intensive) infrastructure to tackle floodingproblems. The goal of this strategy will be to reduce the percentage of people affected byflooding in KMC by targeting the most vulnerable areas and population segments on a prioritybasis.

A number of soft measures, such as a comprehensive approach to planning, proper watershedmanagement, enhancement of disaster management and preparedness, improvement ininstitutional management and accountability, and introduction of sustainable financing arecritical to guarantee the effectiveness of the capital investments in preventing flooding andalso to ensure longer term financial, institutional and environmental sustainability.

Preparation of detailed plans recognizing drainage system complexity and interconnectivityof its elements such as storm water drainage, water supply, wastewater, water pollution control,water reuse, soil erosion, and solid waste management is essential for Kolkata. The planningshould incorporate climate risk factors by clearly spelling out the additional damage arisingout of climate risks and identifying mitigating factors needed in operational plans for keyrelevant agencies.

An integral part of any adaptation strategy in Kolkata would be to support a natural resourcesconservation plan that ensures proper balance of the ecosystem. Investments to maintain andstrengthen culverts and retaining structures and road design that mitigate flooding should be apriority in water logged areas. A proper watershed management would include construction ofstorm water retention infrastructure interlinking ponds and parks in the drainage network toreduce flooding. Land use and building codes should be strengthened to reduce obstructionand encroachments of floodplains and environmentally sensitive areas such as canal banksand wetlands and also to prevent conversion of green spaces and natural areas that can act asretaining zones during flooding to delay runoffs or reduces their volume through infiltration.

Institutional changes are warranted so that (i) the KMC becomes a proficient andautonomous civic body and can operate sewerage and drainage systems on corporate principleswith responsibilities and accountability, (ii) role of the Kolkata Metropolitan DevelopmentAuthority (KMDA), the Kolkata Metropolitan Water and Sanitation Authority (KMWSA) andthe Irrigation and Waterways Department be clearly defined and agreed upon, (iii) maintenance


of storm water flow canals be taken over by the KMC or Department of Municipal Affairs, (iv)private sector participation be introduced for trunk sewer and canal maintenance throughoutthe KMC area, (v) sustainable financing for infrastructure investment and maintenance fromtwo angles - cost reduction and cost recovery- be introduced10, and (vi) widespread adoptionof flood insurance be encouraged.

Investment in hard infrastructure, on the other hand should include (i) de-siltation of thetrunk sewers both in the town and suburban systems to increase the hydraulic capacity and tominimize flooding in the core area of the KMC, (ii) construction of additional trunk sewers inboth the town and suburban systems, (iii) extension of sewerage and drainage facilities in theCossipore - Chitpur area (Borough I) and the areas that were added to the KMC limit in 1984(Boroughs XI to XV), (iv) construction of deep sump pumps and/or rehabilitation of pump aswell as changes to shape the existing sewers (gradation) to improve the hydraulics of the majorsewers approaching the major pumping stations and to increase the flow velocities in the sewersduring flood conditions, (v) an overall improvement in the storm water drainage system fromthe core KMC leading to the Kulti River, and (vi) the renovation and rehabilitation of the canalsystem.

6. CONCLUSION

The results show that urban flooding is a major hazard in Kolkata and can cause severe damagein a number of sectors. Climate change is likely to intensify this problem through a combinationof more intense local precipitation, riverine flooding in the Hooghly and coastal storm surges.In addition, if such intense precipitation is accompanied by extreme weather events such ascyclones, the consequence can be quite serious. By 2050, the annual expected damage in Kolkatais likely to be nearly $5 billion and this is likely to be even higher in future years. This calls forappropriate preventive steps to mitigate the effects of such flooding. Since, many such adaptationstrategies, especially those involving civil construction, tend to have long gestation periods,there is an urgent need to plan ahead to prevent such water-logging in Kolkata.

A major cause of periodic flooding of Kolkata during the rainy season is the adaptationdeficit in sewerage network and treatment infrastructure, drainage infrastructure, and financialresources and institutional capacity to cope with intense precipitation events. These deficitsare likely to intensify in the time horizon of 2050 as the city population rises and with it comesincreasing development and urbanization in a metropolis that has already grown and sprawled,largely in an unplanned way.

Currently, a number of schemes are being implemented in Kolkata to prevent water-logginginvolving a total outlay of $2.7 billion. However, these development plans have a time horizonof only 2025 and also do not account for the possible long-term effects of climate change orany adaptation that may be needed to cope with the problems arising out of climate changeover time (KMPC, 2004). The estimated damage of $5 billion by 2050, once the effect ofclimate changes are included, reveal the need for planning over a longer time frame and alsojustify the need for a much higher level of investment.

The extent of damage from climate change effects found in the study should sensitize allstakeholders about the urgent need for according priority to soft measures involving changes


in policies and functioning of institutions that have been ignored in the past. While suitablesoft measures can minimize the effect of such flooding, there is also a need to make investmentsin hard infrastructure to prevent flooding. The viability of such investments is generallycomputed based on Net Present Value (NPV) and Internal Rates of Return (IRR) criteria usingthe savings in damages. It is likely that once we take into account the additional damage fromclimate change effects many projects not found viable earlier with only current weather datawill become viable. Hence, during any future planning for adaptation in Kolkata, such climatechange effects should be made an integral part in all such NPV and IRR computations.

Acknowledgement

This paper benefited from substantial input, including the hydrological and hydraulic modeling, provided bythe INRM Consults, Delhi, India under the leadership of Dr. Ashvani Gosain and Dr. Sandhya Rao. We aregrateful to Mr. Debal Ray (Department of Environment, Government of West Bengal), Mr. Ardhendu Sen(Chief Secretary, Government of West Bengal), Mr. Sushit K Biswas (Kolkata Environmental ImprovementProject), Mr. Nilangshu Basu (Kolkata Municipal Corporation) and Mr. Samar Ghosh (Secretary Health andFamily Welfare Department, Government of West Bengal) for providing substantial guidance. We are thankfulto Anna C. O’Donnell, Dan Biller, Adriana Damianova, Megumi Muto, John Pethick, Bradfort R. Philips,Neeraj Prasad, Jack Ruitenbeek, Walter Vergara and participants of the stakeholder workshop at Kolkata foruseful comments and suggestions. We would also like to extend our special thanks to Ms. Arati Belle and Ms.Sonia Sandhu for help.

Financial support for this study was provided by Trust Fund for Environmentally Socially SustainableDevelopment supported by Finland and Norway.

The views expressed here are the authors’, and do not necessarily reflect those of the World Bank, its ExecutiveDirectors, the countries they represent, or of the Government of India.

Notes

1. Administrative boundary.

2. This is an estimate of the time interval between two precipitation events of certain intensity. It is a statisticalmeasure of the average recurrence interval over a long period of time and is the inverse of the probabilitythat the event will be exceeded in any one year. Hence, a 100 year precipitation level has a 1% chance ofoccurring in any given year.

3. A1 scenario- World: Convergent world; Economy: Market oriented, rapid economic growth; Population:Peaks in 2050 and then gradually declines; Governance: A convergent world in which regional averageincome per capita converge, current distinction between “poor” and “rich” countries eventually dissolve;A1F1-Energy: Fossil fuels intensive.

4. B1 scenario- World: Convergent world; Economy: service and information based, lower growth than A1;Population: Same as A1; Governance: Global solutions to economic, social and environmental stability;Technology: Clean and resource efficient technologies.

5. Hyetographs are available from the authors upon request.

6. A recent study shows that with an estimated land subsidence of 0.5 meters, the damage caused by waterinundation in KMA by 2070 can be quite devastating (OECD, 2007).

7. This is a commonly adopted practice in such damage assessments as it is difficult to make credibleassumptions about inflation in the future.

8. Source: communication with the Secretary of Department of Health and Family Welfare, Government ofWest Bengal, March 2010.


9. KMC (Roads Department): New Year’s Resolution and Request- Keep Track of Our Progress toward aBetter Kolkata, Today Episode-1. Today we give an account of Roads, Timeframe: January-December,2006.

10. The cost reduction would relate to using appropriate technologies for Kolkata and reducing non-productivecost centers. The cost recovery relates to improving the pricing of service delivery: (i) a shift towardseffective service delivery and associated pricing (including peak and off-peak pricing and flat annual taxfor capital expenditures) and (ii) implementation of appropriate incentives (subsidies, taxation) to encourageadaptive behavior by both public and private entities.

11. The estimate is based on average number of floors in each type of house.

12. Based on 2009 data

References

ACC (2009), ACC Help. http://www.acchelp.in/cost-calculators.asp. Accessed 9 November 2009.

ADB (2008), Climate Change ADB Programs. Strenghthening Mitigation and Adaptation in Asia and thePacific. http://www.donorplatform.org/component/option,com_docman/task,doc_view/gid,940. Accessed21 October 2010.

Chatterjee, R. S., Fruneaub B., Rudantb J., Royc P., Frisonb P., Lakheraa R., Dadhwala V., Saha R. (2006),Subsidence of Kolkata (Calcutta) city, India during the 1990s as observed from space by DifferentialSynthetic Aperture Radar Interferometry (D-InSAR) technique. Remote Sensing Environ 102: 176–185.

ECLAC (2003), Handbook for Estimating the Socio-economic and Environmental Effects of Disasters, http://www.undp.org/cpr/disred/documents/publications/eclac_handbook.pdf. Accessed 12 February 2010.

IMF (2009), World Economic Outlook Database-October 2009. International Monetary Fund, WashingtonDC.

IPCC (2007), Fourth Assessment Report: Climate Change AR4. http://www.ipcc.ch/publications_and_data/publications_and_data_reports.htm#1. Accessed 14 October 2009.

JICA (2008), “Interim Report, Study on Climate Impact Adaptation and Mitigation in Asian Coastal MegaCities” Prepared by the Integrated Research System for Sustainability Science. University of Tokyo, Tokyo.

JNNURM (2005), Status of City Development Plan: Kolkata, Ministry of Urban Development, Government ofIndia http://jnnurm.nic.in/nurmudweb/toolkit/kolkata.htm. Accessed 12 January 2010.

JNNURM (2009), City Development Plan http://www.jnnurmwestbengal.gov.in/HTM/CDP.htm. Accessed 12January 2010.

KMC (2010), Basic Statistics of Kolkata. https://www.kmcgov.in/KMCPortal/jsp/KolkataStatistics.jsp. Accessed14 October 2010.

KMPC (2004), Vision 2025–A Perspective Plan of Kolkata Metropolitan Authority. Kolkata MetropolitanPlanning Committee, Kolkata.

Kok M. (2001), “Damage Functions for the Meuse River Floodplain”, Internal Report. Joint Research Center,European Commission (Ispra), Rome.

Kolkata Municipal Corporation (2007), Kolkata Environmental Improvement Project: Sewerage and Drainagemaster Plan for Kolkata. KMC, Kolkata.

Madheswaran S. (2007), Measuring the Value of Statistacl Life: Estimating Compensating Wage Differentialsamong Workers in India. Social Indicators Research 84(1): 83-96.

Nandy D. (2007), Need for Seismic Microzonation of Kolkata Megacity. Proceedings of Workshop onMicrozonation, Indian Institute of Science, Bangalore, India.


ReDiff Business (2010), GDP: The top 10 cities in India. http://business.rediff.com/slide-show/2010/apr/23/slide-show-1-the-top-10-cities-in-india-by-gdp.htm. Accessed April 21, 2010.

Sugiyama, M. (2008), The Study on Climate Impact Adaptation and Mitigation in Asian Coastal Mega Cities ofIntegrated Research System for Sustainability Science, University of Tokyo, Tokyo.

OECD (2007), Nicholls R., Hanson S., Herweijer C., Patmore N., Hallegatte S., Corfee-Morlot J., Chateau J.,R. Muir-Wood R. Ranking Port Cities with High Exposure and Vulnerability to Climate Extremes: ExposureEstimates. OECD Environment Working Paper 1, ENV/WKP (2007)1, OECD, Paris.

UNFCC (2008), Physical and Socio-economic Trends in Climate-related Risks and Extreme Events, and theirImplications for Sustainable Development http://unfccc.int/resource/docs/2008/tp/03.pdf, Accessed 5August 2010.

UN Department of Economic and Social Affairs (2005), World Urbanization Prospects: The 2005 Revision.UNDESA, New York.

U.S Army Corp of Engineers (2002), HEC-RAS River Analysis System: Hydraulic Reference Manual, November2002.

WHO (2004), The Global Burden of Disease. WHO, Geneva.

World Bank (2010a), Climate Risks and Adaptation in Asian Coastal Megacities: A Synthesis Report. WashingtonDC.

World Bank (2010b), Cities and Climate Change: An Urgent Agenda. Washington DC.

World Bank (2010c), Project Information Document. http://www-wds.worldbank.org/external/default/WDS ContentServer/WDSP/IB/2009/12/15 /000262044_20091215100710/Rendered/PDF/TNRSP010Add0Financial01PID11FINAL.pdf. Accessed 15 April 2010.

World Bank. (2011), Economics of adaptation to Climate Change: Bangladesh http://climatechange.worldbank.org/sites/default/files/documents/EACC_Bangladesh.pdf. Accessed 14 February 2011.

WMO (2010), Current Extreme Weather Events. http://www.wmo.int/pages/mediacentre/news/extremeweathersequence_en.html. Accessed 24 October 2010.

��

Documents

URBAN FLOODING IN A CHANGING CLIMATE: CASE ...serialsjournals.com/serialjournalmanager/pdf/1436348337.pdfUrban Flooding in a Changing Climate: Case Study of Kolkata, India 137 to short,