Chennai is a rapidly urbanizing Indian mega-city and experiences flooding frequently. Literature state that climate change and land use change have a significant impact on the runoff generated every year, making it essential to study the historical trend and forecast changes in land use land cover (LULC) and climate to model runoff. This study considered Adyar watershed for LULC change detection, climate change analysis, and flood forecasting for 2030 and 2050 based on LULC and runoff from 2005 and 2015. A coupled hydrologic–hydraulic model (HEC-HMS and HEC-RAS) was developed to assess flooding for future LULC and climate scenarios. LULC analysis shows an increase in built-up cover by 6%. Climate analysis shows 74% probability of increase in precipitation intensity between 2015 and 2050 compared to 2015. It was observed that depth of flooding increased by 19.4% in 2030 and 60.4% in 2050 compared to 2015. This study makes a structural proposition for flood mitigation through flood carrier canals on the downstream reach of the river which flows through Chennai. The canals were found to significantly reduce overbanking, providing protection against flooding. This is the best measure for providing the highest flood reduction for the study area.

  • Coupled hydrologic and hydraulic model was developed to assess flood inundation of Adyar river basin.

  • Terrset land change modeller was used to project future land use for 2030 and 2050.

  • Climate change projection for 2030 and 2050 was carried out using regional climate models.

  • Flood carrier canals were proposed to mitigate future flooding conditions under extreme events.

  • Developed mitigation measure was analysed through HEC-RAS model.

Flooding usually occurs due to heavy or continuous rainfall resulting in excessive runoff and overflow of river, streams and channels. Flooding has become more of a man-made disaster as urbanization has become the leading cause of most flooding incidents in large cities worldwide (Cao et al. 2020). A combination of changes in rainfall, land use land cover (LULC), and drainage design will increase the vulnerability state of urban areas to pluvial flooding (Willems et al. 2012; Chen et al. 2021).

Land use change simulation aids in understanding the hydrological response of basins under LULC transitions. Recent advancements in computational intelligence and remote sensing help to predict future land use based on historical transition. Many researchers have integrated land change modeller (LCM) with a hydrological model to analyse land use change impact on the hydrological response of basin (Qaiser et al. 2012; Gao et al. 2017, 2020; Woltemade et al. 2020; Aalijahan & Khosravichenar 2021; Ji et al. 2021) Climate change, on the other hand, has increased the intensity of rainfall, which has increased the severity and frequency of flooding (Allan & Soden 2008; Min et al. 2011; Shiu et al. 2012). Average annual precipitation is expected to increase in higher latitudes although the change in precipitation will not be uniform (Aalijahan et al. 2023). This also plays a major role in the flooding of urban areas. Many studies have been carried out on the impact of climate change on runoff (Miller & Hutchins 2017; Kirshen et al. 2018; Didovets et al. 2019). The study of precipitation trends, changes in precipitation due to global warming and prediction of future changes in precipitation is essential for hydrological studies (Aalijahan et al. 2023). Due to its finer spatial resolution regional climate models (RCMs) were considered as an essential tool in regional hydrological impact analysis (Teutschbein et al. 2011). Based on the CMIP5 models, representative concentration pathway (RCP) experiments RCP4.5 and RCP8.5 were widely adopted in several hydrological impact studies (Neupane et al. 2021). Devi et al. (2020) investigated the potential role of retention storage tanks on urban flooding in Chennai in 2050 and proposed dredging to reduce inundation from 1 in 50 year return period flood events. While the previous studies analysed LULC change impact on river runoff, investigation of LULC impact on flood inundation is minimal.

The watershed areas of River Adyar experience flooding almost every year during monsoon. There was extreme flooding on the Adyar flood plain during 1976, 1985, 2005, 2008 and 2015, damaging lives, property and infrastructure (Suriya & Mudgal 2012). It can be stated that this increase in flooding events can be directly related to temporal variations in land use, land cover and climate (Andrés-Doménech et al. 2012; Duvvuri & Balaji 2013; Sarmah & Das 2017; Devi et al. 2019; Das & Md Esraz-Ul-Zannat 2022). Studying these variations and their impact on urban floods will greatly benefit planners and government officials in mitigating floods. The study area needs a permanent solution to address its constant flooding problems. This study is the first of its kind to propose flood carrier canals and analyse its impact on the reduction of flooding in the study area up to 2050 under changing land use and climate.

Study area

The River Adyar is one of the four rivers of the Chennai basin and originates from Chembarambakkam reservoir in the Kanchipuram district. The river collects water from around 200 tanks and flows through Chennai to drain into the Bay of Bengal near Adyar. The total length of the river is 42.5 km and flows through the districts of Chennai, Kanchipuram and Tiruvallur. The catchment area of the River Adyar is around 824 sq.km. The study area falls under the semi-arid tropical climate and receives most of its rainfall from the north east monsoon. Its terrain is mostly flat, with a gentle slope towards the east. Rapid urbanization and industrialization have resulted in the direct disposal of sewage into the river. The 2015 flood that occurred on 2 December when Chennai city received 540 mm of rainfall was catastrophic when, the average rainfall is only 190 mm (Shankar et al. 2017). The observed rainfall event was higher than 1 in 100 years (Devi et al. 2020). The index map of the study area is shown in Figure 1.
Figure 1

Index map of study area.

Figure 1

Index map of study area.

Close modal

This study assesses the impact of LULC change on flooding in the Adyar watershed. The methodology adopted for this study consists of (i) creation of historical LULC maps for the years 2005 and 2015, (ii) prediction of LULC for the years 2030 and 2050, (iii) projection of climate for 2030 and 2050, (iv) rainfall–runoff modelling for 2005, 2015, 2030 and 2050, (v) flood inundation mapping for 2005, 2015, 2030 and 2050, and (vi) design and modelling of canals in downstream reach of the river.

The study area delineation was performed using HEC-HMS 4.9 with Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) digital elevation model (DEM) of 30 m resolution. Historical LULC maps were created using ArcMap 10.8, and LULC maps for 2030 and 2050 were projected using IDRISI TERRSET software. Rainfall data from 1989 to 2019 were collected from Indian Meteorological Department (IMD), Chennai. The detailed methodology to assess impact of LULC and climate change on urban flooding and effect of structural measure on flood mitigation is shown in Figure 2.
Figure 2

Framework to assess LULC change and climate change impact on urban flooding using coupled hydrologic–hydraulic model.

Figure 2

Framework to assess LULC change and climate change impact on urban flooding using coupled hydrologic–hydraulic model.

Close modal
Landsat images were used to create land cover maps. Images from Landsat 5 TM for 2005 and Landsat 8 for 2015 were used to create LULC maps for the respective years. Spectral indices such as normalized difference vegetation index (NDVI), modified normalized difference water index (MNDWI) and normalized difference built-up index (NDBI), and reference images from Google Earth were used for accurate classification of LULC. LULC maps were created under supervised classification using maximum likelihood classification.
formula
(1)
where NIR = near infrared band and red = red band.
NDVI, given in Equation (1), was used in this study to classify the pixels containing vegetation from the Landsat images. This is a step towards creation of land use land cover map for the study area. NDVI measures vegetation by measuring the difference between NIR (strongly reflected by vegetation) and red bands (absorbed by vegetation). The index ranges between −1 and 1, and the positive values representing vegetation and the negative values may indicate built-up land, water and bare soil.
formula
(2)
where green = green band and MIR = middle infrared band.
MNDWI (Xu 2006) uses the green band and MIR to enhance water bodies. The index output ranges between −1 and +1, with positive values representing water bodies. MNDWI classifies the pixels representing water from the Landsat images, which will be used for creating LULC maps for the study area.
formula
(3)
Here, SWIR = shortwave infrared and NIR = near infrared band.

NDBI (Zha et al. 2003) is more accurate in extracting urban built-up land from Landsat images. It uses SWIR and NIR for the extraction of built-up pixels. The index output ranges between −1 and +1, with positive values representing built-up areas.

HEC-HMS 4.9 and HEC-RAS 6.1, widely used in many studies, were used for hydrologic and hydraulic flood modelling. ASTER DEM for river basin delineation, curve number (CN) and rainfall for runoff modelling were given as input into HEC-HMS 4.9. The extent of flooding for precipitation of 2005 and 2015 was mapped using HEC-RAS 6.1. The output hydrographs from HEC-HMS are given as input into HEC-RAS along with Manning's n values, river cross sections and flow details.

For runoff modelling, the loss for each sub-basin was based on the Soil Conservation Service (SCS)-CN values. The soil map of the study area used for the estimation of the CN was obtained from the Institute for Water Studies, Chennai, and the same is shown in Figure 3. The CN was estimated by overlaying the LULC maps of different years with the soil map of the study area. The study area has a combination of clay, clay loam loamy sand, sand, sandy clay, sandy clay loam and sandy loam soils. A significant part of Chennai is made up of clay, and the rest of the soil types form the other two districts of Kanchipuram and Tiruvallur.
Figure 3

Soil map of the River Adyar basin.

Figure 3

Soil map of the River Adyar basin.

Close modal
Based on the historical maps, the land cover maps for 2030 and 2050 were projected using the LCM in the IDRISI TERRSET software. An LCM model is based on the artificial neural network, Markov chain matrices generated by training multilayer perceptron, or logistic regression (Hasan et al. 2020). Population density, slope, elevation, distance to previous disturbance and distance to roads were considered as driving variables for future land use projection (Huang et al. 2017). The driver variables are shown in Figure 4(a)–4(e). Land cover classified images of 1995 and 2005 were used to simulate the land cover for 2015. The accuracy of the simulated image was assessed using the validation module in TERRSET software. Different indices of agreement (KIA) were generated. The Kstandard refers to the simulated image rate of error avoidance, Kno denotes the overall accuracy of the simulation (Chakraborti et al. 2018), Klocation measures accuracy at the grid cell level and KlocationStrata measures how well the grid cell is located in the strata level. The KStandard estimates the model ability in simulating the future land use to acquire perfect classification, and Klocation indicates the locational accuracy of the simulated map. Kno represents the proportion of correctly classified pixels to the expected proportion of correctly classified pixel with respect to location. While simulating land use of categorical maps for future conditions, the location of simulated pixel is as important as the quantity of simulation (Pontius 2000; Ulloa-Espindola & Martin-Fernandez 2021).
Figure 4

Driver variables considered for land change modelling: (a) distance to previous disturbances; (b) distance to roads; (c) slope; (d) elevation; and (e) population density.

Figure 4

Driver variables considered for land change modelling: (a) distance to previous disturbances; (b) distance to roads; (c) slope; (d) elevation; and (e) population density.

Close modal

The simulated image accuracy was classified as poor when k < 0.4, good when k is 0.4–0.75 and excellent when k > 0.7 (Roy et al. 2014).

The changes in these driver variables in the historical maps are projected for obtaining future LULC maps in LCM. Distance to previous disturbance, shown in Figure 4(a), was obtained using the Euclidean distance from the urban transition that had occurred between two satellite images. This driver projects future anthropogenic transition based on existing disturbances. Figure 4(b) shows the distance to roads and depicts the urban growth along the periphery of the study area with an easy commute. The slope of the study area, derived from the DEM using ArcMap 10.8, is shown in Figure 4(c). The percentage of slope varies between 1 and 42. Figure 4(d) shows the elevation map of the study area. The study area has a flat terrain and its elevation varies from 0 to 161 m above mean sea level. Figure 4(e) shows the population density, which is considered the major driving force of urbanization. The population density data were downloaded from ENERGYDATA.INFO.

Climatic scenarios and bias correction

In this study, three RCM projections were acquired from the Coordinated Regional Climate Downscaling Experiment (CORDEX) domain for the historical period 1980–2005 and the projected periods 2006–2030 and 2031–2050 under two emission scenarios RCP4.5 and RCP8.5. The descriptions of global climatic model (GCM) and RCM are shown in Table 1. Every RCM is associated with the systematic errors (biases), which cannot be directly incorporated with hydrological modelling. Bias correction techniques are employed to reduce these uncertainties between the observed and simulated climatic projections using empirical relations. A climate model for hydrological modelling (CMHyd) (Rathjens et al. 2016) was employed to perform bias correction and using linear scaling technique (Teutschbein & Seibert 2012). After bias correction, the base period evaluation of three regional climatic models was carried out using observed precipitation datasets for four stations, namely, Chembarambakkam, Nungambakkam, Chengalpattu and Meenambakkam. The performance metrics such as correlation coefficient (r), root-mean-square error deviation, bias, standard deviation (SD) and Murphy's skill score (ss) were used to evaluate regional climatic model performance. Based on the performance assessment, CNRM-CM5 models have a close resemblance to the observed precipitation than the other two RCMs (Figure 5). Precipitation projected by CNRM-CM5 model was used to perform rainfall-runoff modelling. The study is performed under four scenarios: (i) Scenario 1: 2030 and RCP4.5, (ii) Scenario 2: 2030 and RCP8.5, (iii) Scenario 3: 2050 and RCP 4.5 and (iv) Scenario 4: 2050 and RCP8.5.
Table 1

Description of regional climatic models

ParameterDriving GCMRCMScenariosInstitute
Precipitation CNRM-CM5 RCA4 RCP4.5, RCP8.5 SMHI, Sweden 
ICHEC-EC-EARTH RCA4 RCP4.5, RCP8.5 SMHI, Sweden 
MIROC-MIROC5 RCA4 RCP4.5, RCP8.5 SMHI, Sweden 
ParameterDriving GCMRCMScenariosInstitute
Precipitation CNRM-CM5 RCA4 RCP4.5, RCP8.5 SMHI, Sweden 
ICHEC-EC-EARTH RCA4 RCP4.5, RCP8.5 SMHI, Sweden 
MIROC-MIROC5 RCA4 RCP4.5, RCP8.5 SMHI, Sweden 
Figure 5

Performance assessment of regional climatic models.

Figure 5

Performance assessment of regional climatic models.

Close modal

Land cover analysis

The study area recorded extensive floods in 1998, 2005 and 2015; hence, 2005 and 2015 were considered for land cover analysis and flood modelling in this study. Land cover maps were prepared in ArcMap 10.8 using Google Earth images as a reference for classification. The study area was largely dominated by barren land (41.72% of the study area), followed by agricultural land or shrubs (21.06% of the study area) in 2005. After 2005, the urban land cover continued to increase to 28.54% of the study area by 2015. The period between 2005 and 2015 witnessed a decrease in water bodies (34.741%), forest cover (84.944%), barren land (34.249%) and agriculture and shrubs (57.03%). The rate of increase in the built-up region and decrease in open space and waterbodies indicates rapid urbanization in the catchment, with spatio-temporal changes in land use in recent years. Encroachment along the river banks is also observed, which reduces flow path sustainability (Zope et al. 2016). The LULC maps for 2005 and 2015 are shown in Figure 6(a) and 6(b), respectively.
Figure 6

Historical LULC map of the River Adyar basin for land change modelling: (a) LULC map for 2005 and (b) LULC map for 2015.

Figure 6

Historical LULC map of the River Adyar basin for land change modelling: (a) LULC map for 2005 and (b) LULC map for 2015.

Close modal
IDRISI TERRSET software was used for the prediction of LULC for 2030 and 2050. The LULC maps of 1995 and 2005 were given as input, along with the driver variables mentioned in the previous section. The LULC map of 2015 was then simulated and validated. The Kstandard is 0.7079, Kno is 0.7806, Klocation is 0.7423 and KlocationStrata is 0.7423. Hence, the model has good accuracy and can be accepted for the simulation of future LULC maps (Pontius & Millones 2011). Figure 7 shows the LULC map of 2015 simulated using IDRISI TERRSET software. The change in built-up land cover over the years is shown in Figure 8. The red pixels represent the built-up areas in the study region. The downstream part of the study area is saturated with built-up pixels and movement of urban pixels towards the upstream part can be noted from the figures. It may also be noted that urban expansion is taking place along the river. The detailed explanation of land use change and urbanization is given in the discussion section.
Figure 7

Simulated LULC map for 2015.

Figure 7

Simulated LULC map for 2015.

Close modal
Figure 8

Extent of built-up land cover of the River Adyar basin for the year (a) 2005, (b) 2015, (c) 2030 and (d) 2050.

Figure 8

Extent of built-up land cover of the River Adyar basin for the year (a) 2005, (b) 2015, (c) 2030 and (d) 2050.

Close modal

Curve number

The SCS-CN method calculates effective rainfall based on land cover, antecedent moisture conditions and soil type. The soil map and land cover maps were superimposed in the ArcMap 10.8 environment to generate the CN map of the study area. The CN obtained is given as input in the HEC-HMS environment to determine the losses and estimate the runoff for each precipitation event.

The CN map is generated for 2005, 2015, 2030 and 2050. The soil map was analysed, and the soil types were classified into different hydrologic groups based on the type of soil (Table 2). The LULC map was superimposed on the soil map to obtain the CN for each pixel of the study area. The CN map was then obtained by calculating the composite CN of each sub-basin of the study area.

Table 2

Soil group classification based on land use

Soil groupMinimum infiltration rate (mm/s)Soil typeLand use
UrbanAgricultureShrubsForestWaterWetlandsBarren
0.0021–0.0032 High infiltration 87 67 55 30 77 
0.0011–0.0021 Moderate infiltration 87 78 56 55 86 
0.00035–0.0011 Slow infiltration 87 85 70 70 91 
0–0.00035 Very slow 87 89 77 77 94 
Soil groupMinimum infiltration rate (mm/s)Soil typeLand use
UrbanAgricultureShrubsForestWaterWetlandsBarren
0.0021–0.0032 High infiltration 87 67 55 30 77 
0.0011–0.0021 Moderate infiltration 87 78 56 55 86 
0.00035–0.0011 Slow infiltration 87 85 70 70 91 
0–0.00035 Very slow 87 89 77 77 94 

It is observed that the CN is gradually increasing from 2005 to 2050. CN increases with an increase in impervious land cover, thus proving the increase in urban land cover in the study area. The CN values for different years are shown in Table 3.

Table 3

Curve numbers for 2005, 2015, 2030 and 2050

Basin nameCN 2005CN 2015CN 2030% Increase in CN from the base year 2005CN 2050% Increase in CN from the base year 2005
W620 85.85 90.06 91.00 6.00 91.51 6.60 
W630 83.90 91.92 93.97 12.01 95.12 13.37 
W650 87.02 90.02 95.26 9.46 96.32 10.69 
W740 87.01 90.74 93.19 7.10 95.32 9.55 
W750 89.52 91.35 92.51 3.34 93.05 3.95 
W770 87.45 88.89 89.22 2.02 91.06 4.12 
W860 88.70 89.31 89.56 0.97 94.01 5.99 
W920 86.92 89.20 89.33 2.77 93.45 7.52 
W970 82.69 89.29 90.88 9.91 91.07 10.14 
W980 88.00 88.39 89.47 1.67 94.30 7.15 
W1040 88.21 88.21 88.48 0.31 91.17 3.36 
W1050 84.25 84.25 84.69 0.52 88.38 4.90 
W1080 86.61 89.29 89.81 3.70 91.38 5.51 
W1090 88.10 90.00 91.16 3.48 94.54 7.32 
Basin nameCN 2005CN 2015CN 2030% Increase in CN from the base year 2005CN 2050% Increase in CN from the base year 2005
W620 85.85 90.06 91.00 6.00 91.51 6.60 
W630 83.90 91.92 93.97 12.01 95.12 13.37 
W650 87.02 90.02 95.26 9.46 96.32 10.69 
W740 87.01 90.74 93.19 7.10 95.32 9.55 
W750 89.52 91.35 92.51 3.34 93.05 3.95 
W770 87.45 88.89 89.22 2.02 91.06 4.12 
W860 88.70 89.31 89.56 0.97 94.01 5.99 
W920 86.92 89.20 89.33 2.77 93.45 7.52 
W970 82.69 89.29 90.88 9.91 91.07 10.14 
W980 88.00 88.39 89.47 1.67 94.30 7.15 
W1040 88.21 88.21 88.48 0.31 91.17 3.36 
W1050 84.25 84.25 84.69 0.52 88.38 4.90 
W1080 86.61 89.29 89.81 3.70 91.38 5.51 
W1090 88.10 90.00 91.16 3.48 94.54 7.32 

Projected 1-day precipitation

Three climatic models, namely, ICHEC-EC-EARTH, MIROC-MIROC5 and CNRM-CERFACS-CNRM-CM5, were compared in the MATLAB programming language. The comparison showed that the results of CNRM-CERFACS-CNRM-CM5 were closer to the observed historical values of precipitation when compared to the other models. Hence, the CNRM-CERFACS-CNRM-CM5 model was used for the projection of precipitation for RCP4.5 and RCP8.5 scenarios for two time periods, namely, near future (2020–2035) and far future (2035–2050). The maximum 1-day precipitation that can occur in the future is shown in Table 4 for near future (2006–2030) and far future (2031–2050) periods.

Table 4

Projected daily precipitation

ScenarioPeriodChembarambakkam (mm)Nungambakkam (mm)Chengalpattu (mm)Meenambakkam (mm)
RCP4.5 2006–2030 389.14 378.11 190.23 370.21 
RCP8.5 422.67 393.45 217.59 398.21 
RCP4.5 2031–2050 527.03 638.64 414.36 569.88 
RCP8.5 642.32 623.45 457.39 607.46 
ScenarioPeriodChembarambakkam (mm)Nungambakkam (mm)Chengalpattu (mm)Meenambakkam (mm)
RCP4.5 2006–2030 389.14 378.11 190.23 370.21 
RCP8.5 422.67 393.45 217.59 398.21 
RCP4.5 2031–2050 527.03 638.64 414.36 569.88 
RCP8.5 642.32 623.45 457.39 607.46 

Rainfall–runoff modelling

The impact of LULC change on flooding was studied through rainfall–runoff modelling. The changes in the flow of river for changes in LULC were modelled through this process. Rainfall–runoff modelling requires the depth of rainfall, CN and DEM, and many models are available for rainfall–runoff modelling. This study uses HEC-HMS 4.9, which has been widely used and accepted (Meenu et al. 2013).

The HEC-HMS model was simulated and validated for the 2015 floods in Chennai. The discharge details for Chembarambakkam reservoir were collected from Public Works Department, Government of Tamil Nadu. The calibration metrics of the model were SD = 0.3, percent bias = 7.84% and Nash–Sutcliffe efficiency = 0.881, which show that the model is capable of performing with greater accuracy. Figure 9 shows the validation results for 2015. After validation, the results for 2030 and 2050 were simulated in HEC-HMS. The daily rainfall values were converted into hourly based on the IMD one-third rule, which is
formula
(4)
formula
(5)
where Rt = rainfall depth in mm for durations less than 24 h, R24 is the daily rainfall depth in mm, t = required duration in hours, n = 1/3, and I = rainfall intensity in mm/h.
Figure 9

Validation of the HEC-HMS model.

Figure 9

Validation of the HEC-HMS model.

Close modal
Figure 10 shows the simulated outflow for 2030 and 2050. The figure shows the simulated peak discharge for 2030 and 2050 to be 2,457.7 and 2,500 m3/s, respectively. Also, the peak in 2050 is attained earlier than in 2030, which is an impact of change in LULC and climate.
Figure 10

Simulation of rainfall – runoff of the River Adyar basin: (a) 2030 LULC and (b) 2050 LULC.

Figure 10

Simulation of rainfall – runoff of the River Adyar basin: (a) 2030 LULC and (b) 2050 LULC.

Close modal

Flood modelling

After the rainfall–runoff modelling was performed, the extent of flood inundation in the study area was mapped. Flood inundation mapping requires the hydrograph obtained from the rainfall–runoff model as input. Among the different models available for inundation mapping, HEC-RAS 6.1 was selected for this study (Zhang et al. 2022).

Manning's n value was assigned as 0.03 for canals, and steady flow analysis was performed. The HEC-RAS model was validated with the observed field depths, and the model showed good accuracy with R2 = 0.98. The model was hybridized with Google Earth to view the spatial extent of flooding across the River Adyar basin. The depth of flow in the River Adyar for 2015, 2030 and 2050 for the four scenarios is shown in Figure 11, and the runoff values are given in Table 5.
Table 5

Outflow for different LULC and climate scenarios

YearClimate modelOutflow in m3/s
ChembarambakkamChengalpattuMeenambakkam
2030 RCP 4.5 666.83 980.41 1831 
2030 RCP 8.5 727.92 1100.24 2022 
2050 RCP 4.5 954.98 1686.39 3109.01 
2050 RCP 8.5 1164.90 2096.81 3552 
YearClimate modelOutflow in m3/s
ChembarambakkamChengalpattuMeenambakkam
2030 RCP 4.5 666.83 980.41 1831 
2030 RCP 8.5 727.92 1100.24 2022 
2050 RCP 4.5 954.98 1686.39 3109.01 
2050 RCP 8.5 1164.90 2096.81 3552 
Figure 11

Depth of flow in river for different climatic scenarios.

Figure 11

Depth of flow in river for different climatic scenarios.

Close modal
It can be inferred from Table 5 that the outflow has increased in all scenarios in Chembarambakkam, whereas in the case of Chengalpattu and Meenambakkam stations, the outflow increased only for 2050 scenario. The difference between the outflow from each scenario and the outflow from 2015 is tabulated in Table 6. The runoff at Chembarambakkam, Chengalpattu and Meenambakkam is shown in Figure 12.
Table 6

Difference between outflow of different scenarios and base period







 






 
Figure 12

Outflow from rain gauge stations for climate and land use scenarios.

Figure 12

Outflow from rain gauge stations for climate and land use scenarios.

Close modal

Flood inundation mapping

The inundation map for 2015 obtained from HEC-RAS 6.1 is shown in Figure 13. Since most of the precipitation values of 2030 and 2050 are higher than that in 2015, the extent of inundation will also be higher.
Figure 13

Flood inundation map for 2015.

Figure 13

Flood inundation map for 2015.

Close modal

Flood carrier canals

There are many previous studies (Odemerho 2014; Chen et al. 2015; Padmanaban et al. 2017; Cao et al. 2020) on flood inundation mapping and the hazards related to the same. The mitigation measures suggested in these studies were either maintenance or policy-related suggestions to the government or policymakers. This study proposes the design of canals on either side of the downstream reach of the River Adyar as a permanent solution for the constant flooding in the watershed.

Flood carrier canals are designed in the downstream reach of the river, the part of the river running within Chennai, as that part of the river floods almost every year. The canals are designed to transport the excess water that overtops the river banks to the outlet at a velocity higher than that in the river.

The flood carrier canal starts from Anagaputhur, the place where the river enters Chennai, up to Kottur, near the outlet of the river. Figure 14 shows a schematic diagram of the canal. Several trials were performed in HEC-RAS software by changing the dimensions of the canal, reach length and its slope to determine the dimensions of the canals. The optimized dimensions, reach length and slope were selected based on the highest reduction in flood achieved during the trials. The canal runs to a length of 25 km, and the width of the canal was fixed as 10 m as this width provides the maximum mitigation of flood. The canal is designed with a slope of 1 in 3.85 m, whereas the slope in the river is around 1 in 81.9 m. Figure 15(a) shows the plan of the entire stretch of the flood carrier canal. To have a closer view of the plan, Figure 15(b) shows a part of the canal, highlighted between the two lines in Figure 15(a).
Figure 14

Schematic diagram of flood carrier canals.

Figure 14

Schematic diagram of flood carrier canals.

Close modal
Figure 15

Flood carrier canal design proposed downstream of the River Adyar basin: (a) plan of flood carrier canal and (b) plan of highlighted part of flood carrier canal.

Figure 15

Flood carrier canal design proposed downstream of the River Adyar basin: (a) plan of flood carrier canal and (b) plan of highlighted part of flood carrier canal.

Close modal

LULC change analysis

The LULC maps of 2030 and 2050, simulated in the IDRISI TERRSET software using the classified LULC maps of 2005 and 2015, show a steady increase in the built-up and barren land cover classes until 2050 (Figure 16(a) and 16(b)). Table 7 and Figure 8 show the percentage of area under each land cover class for 2005, 2015, 2030 and 2050. Table 7 shows that the area under the urban class has increased from 28.54% of the total study area in 2015 to 31.58% in 2030 and 33.65% in 2050. The area under the barren land cover class has also increased from 41.44% of the total study area in 2015 to 46.32% in 2030 and 48.55% in 2050. It is also evident that urbanization in the study area is mainly concentrated within the Chennai limits around the airport, along the south and around existing water bodies, which has led to their shrinkage and closure.
Table 7

Percentage of area under each LULC for 2005, 2015, 2030 and 2050

LULCPercentage of total area for different years
2005201520302050
Urban 15.59 28.54 31.58 33.65 
Water 15.07 14.86 9.52 4.99 
Forest 6.55 1.49 1.13 1.26 
Barren 41.72 41.44 46.32 48.55 
Shrubs and agriculture 21.03 13.67 11.44 11.55 
LULCPercentage of total area for different years
2005201520302050
Urban 15.59 28.54 31.58 33.65 
Water 15.07 14.86 9.52 4.99 
Forest 6.55 1.49 1.13 1.26 
Barren 41.72 41.44 46.32 48.55 
Shrubs and agriculture 21.03 13.67 11.44 11.55 
Figure 16

Simulated LULC map of the River Adyar basin using land change modeller for (a) 2030 and (b) 2050.

Figure 16

Simulated LULC map of the River Adyar basin using land change modeller for (a) 2030 and (b) 2050.

Close modal

Climate change analysis

The rainfall that occurred in 2015 was assessed as a 125-year return period rainfall with maximum rainfall at Chembarambakkam, Nungambakkam and Meenambakkam stations. It is globally recognised that, as a result of climate change, the intensity of rainfall may increase (Markus et al. 2016; Giorgi et al. 2019; Myhre et al. 2019). The probability of occurrence of precipitation greater than the 2015 event is 49% over the next 10 years, 74% in the next 20 years and 87% in the next 30 years, making the study area more prone to flooding. Figure 17 shows the maximum 1-day precipitation values predicted in the four scenarios.
Figure 17

Maximum 1-day rainfall for different scenarios.

Figure 17

Maximum 1-day rainfall for different scenarios.

Close modal

Runoff analysis

It was observed from the results that due to changes in LULC and climate between 2015 and 2030 and 2030 and 2050, the peak discharge in Chembarambakkam increased and the peak is attained earlier in 2030 and 2050 than in 2015. The peak discharge in 2015 was 2,198.5 m3/s, whereas the peak discharge in 2030 and 2050 was simulated to be 2,457.7 and 2,500 m3/s, respectively. Hence, the impact of change in LULC and climate can be clearly understood from the aforementioned results.

Green infrastructure for flood mitigation

Nature-based solutions such as artificial recharge, rainwater harvesting and green roofs help in augmenting groundwater and thereby reducing flooding. Rooftop rainwater harvesting has been made mandatory in the state of Tamil Nadu since 2001 when Chennai faced acute water crisis. On the other hand, the government undertook the restoration work of the Adyar Creek in 2010 which increased the water spread area from 5.53 to 59% (CRRT 2020). This reduced sewage outflow into the river, thereby improving vegetation and aquatic life. But this did not have much effect for an extreme event like the 2015 floods. In addition, the Greater Chennai Corporation houses 894 km of stormwater drains and 30 canals. The drains have in situ rainwater harvesting facilities at intervals of 30 m to enhance groundwater recharge. These drains and canals can be effectively used to reduce the flood by providing recharge shafts along the stormwater drain/canal beds at a more frequent intervals of 10 m.

Recharge zones were identified in the study area by overlaying LULC map and soil map. The areas under forest, barren and shrubs with sandy or loamy soils were identified as recharge zones, and the same is shown in Figure 18. Recharge pits can be provided in the recharge zones to enhance groundwater recharge. The study area is capable of accommodating around 20 recharge pits, thereby recharging groundwater.
Figure 18

Suitable sites for recharge zones.

Figure 18

Suitable sites for recharge zones.

Close modal

It can be noted that the recharge zones are concentrated upstream of the basin as the downstream side is mostly urbanized. The tanks and rainwater harvesting existing in the city are not sufficient enough to mitigate inundation from extreme climatic events. Therefore, the downstream city part of the study area, which was the most affected in 2015, requires an alternative solution, other than green infrastructures, for better flood mitigation.

Structural measure for flood mitigation

It was observed that the highest depth of flooding in 2015 was 10.31 m and occurred in Tharapakkam road. To understand the effect of LULC on depth of flooding, modelling was performed using 2015 precipitation and the LULC maps for 2030 and 2050. The depth of flooding was estimated as 13.11 m for 2030 and 13.4 m for 2050 for constant precipitation and changing LULC. The depth of flooding increases by 2.81 m between 2015 and 2030 and by 0.29 m between 2030 and 2050 proving the impact of LULC change on the runoff pattern. The effect of the canals on the flood mitigation was studied for the 2015 flood. The maximum water level in the river during the 2015 flood reduced from 18.19 to 3.42 m in the presence of flood carrier canals on the downstream reach. This decrease in the depth of flow is due to the reason that the excess flow in the river is transported at a higher velocity to the outlet through the canals. The reduction of flood inundation by the canals for the 2015 flood event is shown in Figure 19.
Figure 19

Impact of flood carrier canal on flood inundation depth: (a) observed flood inundation without canal and (b) simulated flood inundation with canal.

Figure 19

Impact of flood carrier canal on flood inundation depth: (a) observed flood inundation without canal and (b) simulated flood inundation with canal.

Close modal

Scenario 1: The hydraulic depth of flow in the river is 11.97 m. Although the volume of precipitation is less in this scenario than in 2015, it can be noted that the hydraulic depth of flow in the river increases from 10.31 m in 2015 to 11.97 m in 2030. Total discharge in the river increases from 1,344.1 m3/s in 2015 to 1,837 m3/s in 2030. This increase in the depth can be attributed to the change in the landuse and climate, which are in agreement with the previous studies (Akter et al. 2018; Hung et al. 2020).

Scenario 2: The hydraulic depth of flow is 12.31 m in the river. The total discharge in the river increases to 2,022 m3/s. The precipitation is higher by 12.46% only in Meenambakkam compared to 2015, resulting in 9.6% increase in runoff.

The hydraulic depths of the flow in the left and right canals are 8.03 and 8.8 m, respectively, for scenario 1 and 9.6 and 10.62 m, respectively, for scenario 2. The presence of these flood carrier canals reduces the depth of flow in the river to a maximum of 3.92 m. Figure 20(a) shows the inundation map for scenario 1, and Figure 20(b) shows the inundation map for scenario 2. The figures clearly show the reduction of the depth of flow in the downstream reach of the river, where the canals are designed.
Figure 20

Flood inundation map for different climate and LULC scenarios: (a) flood inundation map for 2030 for RCP4.5; (b) flood inundation map for 2030 for RCP8.5; (c) flood inundation map for 2050 for RCP4.5; and (d) flood inundation map for 2050 for RCP8.5.

Figure 20

Flood inundation map for different climate and LULC scenarios: (a) flood inundation map for 2030 for RCP4.5; (b) flood inundation map for 2030 for RCP8.5; (c) flood inundation map for 2050 for RCP4.5; and (d) flood inundation map for 2050 for RCP8.5.

Close modal

Scenario 3: An increase in precipitation results in 59.5% increase in runoff, which should be accommodated and safely transported to the outlet without flooding. The hydraulic depth of flow is 15.44 m in the river.

Scenario 4: The study area will witness a 70.1% increase in runoff. These results are similar to the increase in urban flooding for the RCP8.5 scenarios in China (52%) (Zhou et al. 2018) and the United States (Hettiarachchi et al. 2018) (10–170%). The hydraulic depth of flow is 16.54 m in the river.

The hydraulic depths of flow in the left and right canals are 12.51 and 13.9 m, respectively, for scenario 3 and 13.87 and 15.42 m, respectively, for scenario 4. The depth of flow in the river reduces to 4.01 m in the presence of canals and overbank flow reduces to zero. Figure 20(c) shows the inundation map for scenario 3, and Figure 20(d) shows the inundation map for scenario 4. The figures show the huge impact of the canals on flood reduction for the aforementioned scenarios. The depth of flow in the canals in all the scenarios is well below the top level, and hence, they make the watershed safe against flooding due to changing land use and climate change up to 2050.

Chennai is one of the megacities in India that faces flooding almost every monsoon. In addition to its flat terrain, increasing imperviousness due to changes in LULC and increasing intensity of precipitation due to climate change result in higher floods in the city.

The built-up land cover will increase from 15.59% of the total area of the Adyar watershed in 2005 to 31.58% in 2030 and to 33.65% in 2050. The other land cover classes such as waterbodies, forest, barren land, shrubs and agricultural lands will decrease, proving that these land cover classes are converted to built-up land due to urbanization. The climate change analysis shows that precipitation intensity is likely to increase in the future. It is observed from the hydrologic and hydraulic modelling of floods in the study area that the changes in LULC and climate will result in a 19.4% increase in the depth of flooding in 2030 and a 60.43% increase in the depth of flooding in 2050 compared to 2015. The depth of flow increases by 211.5% between 2030 and 2050. Hence, it is important to predict the temporal changes in LULC and climate to determine the increase in runoff that could be expected in the study area.

This study attempts a structural measure to mitigate floods through the design of canals on either side of the downstream reach of the River Adyar. This structural measure is a first of its kind for Chennai and has not been addressed in any of the previous studies. While studies in the past have suggested artificial recharge and policy-related measures for flood mitigation, this study analyses the impact of flood carrier canals along the river. This study also proves that this structural measure provides maximum efficiency in reducing the discharge in the river under extreme rainfall events.

Chennai, with a history of 383 years, with urbanization nearing saturation, is known to be a flat terrain. Hence, land acquisition for the construction of flood-carrying canals may be viewed as a constraint. However, considering the high efficiency of the canals compared to the measures proposed in the previous studies, the policymakers may consider this measure as a technically viable option for flood prevention under future LULC and climate scenarios. The structural requirements for construction and the economic feasibility of this canal are not considered here as this study concentrates on hydrologic and hydraulic modelling. The feasibility of the construction of canals in the downstream part of the study area can be undertaken as the future scope of this study. The study considers the empirical SCS-CN method for hydrological modelling, which has its limitations (Verma et al. 2017). The limitation associated with the coarser resolution of DEM also plays a major role in hydraulic modelling (Vozinaki et al. 2016). The projected land use is hypothetical in nature, which may alter the simulated runoff in future conditions. However, the simulated runoff and the proposed structural measures aid the policymakers in avoiding future disastrous events.

The authors are thankful to Institute for Water Studies, Chennai, and Indian Meteorological Department (IMD, Chennai) for providing necessary data to carry out this research work.

N.A. Sreemanthrarupini: conceptualization, methodology, software, and writing – original draft. R. Saravanan: conceptualization, methodology, validation, and writing – review & editing. R. Balamurugan: formal analysis and software. L. Balaji: software, validation, formal analysis, and writing – review & editing.

This work was supported by Anna University, Chennai under Anna Centenary Research Fellowship (ACRF) scheme (Grant No: CFR/ACRF/2017/44).

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.

Aalijahan
M.
&
Khosravichenar
A.
2021
A multimethod analysis for average annual precipitation mapping in the Khorasan Razavi province (Northeastern Iran)
.
Atmosphere
12
(
5
),
592
,
1–19. https://doi.org/10.3390/atmos12050592
.
Aalijahan
M.
,
Karatas
A.
,
Lupo
A. R.
,
Efe
B.
&
Khosravichenar
A.
2023
Analyzing and modeling the spatio-temporal changes and the impact of GLOTI index on precipitation in the Marmara region of Turkiye
.
Atmosphere
14
(
3
),
489
,
1–25. https://doi.org/10.3390/atmos14030489
.
Akter
T.
,
Quevauviller
P.
,
Eisenreich
S. J.
&
Vaes
G.
2018
Impacts of climate and land use changes on flood risk management for the Schijn River, Belgium
.
Environmental Science & Policy
89
(
2018
),
163
175
.
Andrés-Doménech
I.
,
Montanari
A.
&
Marco
J. B.
2012
Efficiency of storm detention tanks for urban drainage systems under climate variability
.
Journal of Water Resources Planning and Management
138
(
1
),
36
46
.
https://doi.org/10.1061/(asce)wr.1943-5452.0000144
.
Cao
R.
,
Li
F.
&
Feng
P.
2020
Exploring the hydrologic response to the urban building coverage ratio by model simulation
.
Theoretical and Applied Climatology
140
(
3–4
),
1005
1015
.
https://doi.org/10.1007/s00704-020-03139-x
.
Chakraborti
S.
,
Das
D. N.
,
Mondal
B.
,
Shafizadeh-Moghadam
H.
&
Feng
Y.
2018
A neural network and landscape metrics to propose a flexible urban growth boundary: A case study
.
Ecological Indicators
93
,
952
965
.
https://doi.org/10.1016/j.ecolind.2018.05.036
.
Chen
Y.
,
Zhou
H.
,
Zhang
H.
,
Du
G.
&
Zhou
J.
2015
Urban flood risk warning under rapid urbanization
.
Environmental Research
139
,
3
10
.
https://doi.org/10.1016/j.envres.2015.02.028
.
Chennai Rivers Restoration Trust
2020
Government of Tamil Nadu
.
Available from: https://www.chennairivers.gov.in/ (accessed 23 November 2022).
Devi
N.
,
Sridharan
B.
&
Kuiry
S. N.
2019
Impact of urban sprawl on future flooding in Chennai city, India
.
Journal of Hydrology
574
,
486
496
.
https://doi.org/10.1016/j.jhydrol.2019.04.041
.
Devi
N.
,
Sridharan
B.
,
Bindhu
V. M.
,
Narasimhan
B.
,
Bhallamudi
S. M.
,
Bhatt
C. M.
,
Usha
T.
,
Vasan
D. T.
&
Kuiry
S. N.
2020
Investigation of role of retention storage in tanks (small water bodies) on future urban flooding: A case study of Chennai City, India
.
Water
12
(
10
),
2875
.
https://doi.org/10.3390/w12102875
.
Didovets
I.
,
Kryanova
V.
,
Burger
G.
,
Snizhko
S.
,
Balabukh
V.
&
Bronstert
A.
2019
Climate change impact on regional floods in the Carpathian Region
.
Journal of Hydrology: Regional Studies
22
,
1
14
.
https://doi.org/10.1016/j.ejrh.2019.01.002
.
Duvvuri
S.
&
Balaji
N.
2013
Flood inundation mapping of Thamiraparani River Basin using HEC-Geo RAS and SWAT
.
International Journal Of Engineering Research & Technology
2
(
7
),
1408
1420
.
Gao
Y.
,
Yuan
Y.
,
Wang
H.
,
Schmidt
A. R.
,
Wang
K.
&
Ye
L.
2017
Examining the effects of urban agglomeration polders on flood events in Qinhuai River basin, China with HEC-HMS model
.
Water Science and Technology
75
(
9
),
2130
2138
.
Gao
Y.
,
Chen
J.
,
Luo
H.
&
Wang
H.
2020
Prediction of hydrological responses to land use change
.
Science of the Total Environment
708
,
134998
.
https://doi.org/10.1016/j.scitotenv.2019.134998
.
Giorgi
F.
,
Raffaele
F.
&
Coppola
E.
2019
The response of precipitation characteristics to global warming from climate projections
.
Earth System Dynamics
10
(
1
),
73
89
.
https://doi.org/10.5194/esd-10-73-2019
.
Hasan
S.
,
Shi
W.
,
Zhu
X.
,
Abbas
S.
&
Khan
H. U. A.
2020
Future simulation of land use changes in rapidly urbanizing south China based on land change modeler and remote sensing data
.
Sustainability
12
(
11
),
4350
.
https://doi.org/10.3390/su12114350
.
Hettiarachchi
S.
,
Wasko
C.
&
Sharma
A.
2018
Increase in flood risk resulting from climate change in a developed urban watershed – The role of storm temporal patterns
.
Hydrology and Earth System Sciences
22
(
3
),
2041
2056
.
Huang
L.
,
Zeng
G.
,
Liang
J.
,
Hua
S.
,
Yuan
Y.
,
Li
X.
,
Dong
H.
,
Liu
J.
,
Nie
S.
&
Liu
J.
2017
Combined impacts of land use and climate change in the modeling of future groundwater vulnerability
.
Journal of Hydrologic Engineering.
22
,
05017007-1
05017007-11
.
https://doi.org/10.1061/(ASCE)HE.1943-5584.0001493
.
Hung
C.-L. J.
,
James
L. A.
,
Carbone
G. J.
&
William
J. M.
2020
Impacts of combined land-use and climate change on streamflow in two nested catchments in the Southeastern United States
.
Ecological Engineering
143
,
105665
.
Ji
G.
,
Lai
Z.
,
Xia
H.
,
Liu
H.
&
Wang
Z.
2021
Future runoff variation and flood disaster prediction of the Yellow River Basin based on CA-Markov and SWAT
.
Land
10
(
4
),
421
.
https://doi.org/10.3390/land10040421
.
Kirshen
P.
,
Aytur
S.
,
Hecht
J.
,
Walker
A.
,
Burdick
D.
,
Jones
S.
,
Fennessey
N.
,
Bourdeau
R.
&
Mather
L.
2018
Integrated urban water management applied to adaptation to climate change
.
Urban Climate
24
,
247
263
.
https://doi.org/10.1016/j.uclim.2018.03.005
.
Markus
G. D.
,
Andrew
L. L.
,
Lisa
V. A.
,
Paul
G. A.
&
Nicola
M.
2016
More extreme precipitation in the world's dry and wet regions
.
Nature Climate Change
6
(
5
),
508
513
.
https://doi.org/10.1038/nclimate2941
.
Meenu
R.
,
Rehana
S.
&
Mujumdar
P. P.
2013
Assessment of hydrologic impacts of climate change in Tunga-Bhadra river basin, India with HEC-HMS and SDSM
.
Hydrological Processes
27
(
11
),
1572
1589
.
https://doi.org/10.1002/hyp.9220
.
Miller
J. D.
&
Hutchins
M.
2017
The impacts of urbanisation and climate change on urban flooding and urban water quality: A review of the evidence concerning the United Kingdom
.
Journal of Hydrology: Regional Studies
12
,
345
362
.
https://doi.org/10.1016/j.ejrh.2017.06.006
.
Min
S. K.
,
Zhang
X.
,
Zwiers
F. W.
&
Hegerl
G. C.
2011
Human contribution to more-intense precipitation extremes
.
Nature
470
,
378
381
.
Myhre
G.
,
Alterskjær
K.
,
Stjern
C. W.
,
Hodnebrog
Ø.
,
Marelle
L.
,
Samset
B. H.
,
Sillmann
J.
,
Schaller
N.
,
Fischer
E.
,
Schulz
M.
&
Stohl
A.
2019
Frequency of extreme precipitation increases extensively with event rareness under global warming
.
Scientific Reports
9
(
1
),
1
10
.
https://doi.org/10.1038/s41598-019-52277-4
.
Neupane
S.
,
Shrestha
S.
,
Ghimire
U.
,
Mohanasundaram
S.
&
Ninsawat
S.
2021
Evaluation of the CORDEX regional climate models (RCMs) for simulating climate extremes in the Asian cities
.
Science of the Total Environment
797
,
149137
.
https://doi.org/10.1016/j.scitotenv.2021.149137
.
Odemerho
F. O.
2014
Building climate change resilience through bottom-up adaptation to flood risk in Warri, Nigeria
.
Environment and Urbanization
27
(
1
),
139
160
.
https://doi.org/10.1177/0956247814558194
.
Padmanaban
R.
,
Bhowmik
A. K.
,
Cabral
P.
,
Zamyatin
A.
,
Almegdadi
O.
&
Wang
S.
2017
Modelling urban sprawl using remotely sensed data: A case study of Chennai City, Tamilnadu
.
Entropy
19
(
4
),
163
.
https://doi.org/10.3390/e19040163
.
Pontius
R. G.
2000
Quantification error versus location error in comparison of categorical maps
.
Photogrammetric Engineering and Remote Sensing
66
(
8
),
1011
1016
.
Pontius
R. G.
&
Millones
M.
2011
Death to Kappa: Birth of quantity disagreement and allocation disagreement for accuracy assessment
.
International Journal of Remote Sensing
32
(
15
),
4407
4429
.
https://doi.org/10.1080/01431161.2011.552923
.
Rathjens
H.
,
Bieger
K.
,
Srinivasan
R.
,
Chaubey
I.
&
Arnold
J. G.
2016
CMhyd User Manual
.
Roy
H. G.
,
Fox
D. M.
&
Emsellem
K.
2014
Predicting land cover change in a Mediterranean catchment at different time scales
.
Computational Science and Its Applications – ICCSA
2014
,
315
330
.
https://doi.org/10.1007/978-3-319-09147-1_23
.
Sarmah
T.
&
Das
S.
2017
Urban flood mitigation planning for Guwahati: A case of Bharalu basin
.
Journal of Environmental Management
206
,
1155
1165
.
https://doi.org/10.1016/j.jenvman.2017.10.079
.
Shankar
C. P.
,
Shanmugam
M.
,
Pradeep
C.
&
Thirumalaivasan
D.
2017
Flood modeling for a part of Adyar river basin using HEC-HMS and HEC-RAS for the December 2015 floods
.
Journal of Ecology Environmental and Conservation.
24
,
260
263
.
Shiu
C. J.
,
Liu
S. C.
,
Fu
C.
,
Dai
A.
&
Sun
Y.
2012
How much do precipitation extremes change in a warming climate?
Geophysical Research Letters
39
,
1
5
.
Suriya
S.
&
Mudgal
B. V.
2012
Impact of urbanization on flooding: The Thirusoolam sub watershed – A case study
.
Journal of Hydrology
412
,
210
219
.
https://doi.org/10.1016/j.jhydrol.2011.05.008
.
Teutschbein
C.
&
Seibert
J.
2012
Bias correction of regional climate model simulations for hydrological climate-change impact studies: Review and evaluation of different methods
.
Journal of Hydrology
456–457
,
12
29
.
https://doi.org/10.1016/j.jhydrol.2012.05.052
.
Teutschbein
C.
,
Wetterhall
F.
&
Seibert
J.
2011
Evaluation of different downscaling techniques for hydrological climate-change impact studies at the catchment scale
.
Climate Dynamics
37
(
9–10
),
2087
2105
.
https://doi.org/10.1007/s00382-010-0979-8
.
Verma
S.
,
Verma
R. K.
,
Mishra
S. K.
,
Singh
A.
&
Jayaraj
G. K.
2017
A revisit of NRCS-CN inspired models coupled with RS and GIS for runoff estimation
.
Hydrological Sciences Journal
62
(
12
),
1891
1930
.
https://doi.org/10.1080/02626667.2017.1334166
.
Vozinaki
A. E. K.
,
Morianou
G. G.
,
Alexakis
D. D.
&
Tsanis
I. K.
2016
Comparing 1D and combined 1D/2D hydraulic simulations using high-resolution topographic data: A case study of the Koiliaris basin, Greece
.
Hydrological Sciences Journal
62
(
4
),
642
656
.
https://doi.org/10.1080/02626667.2016.1255746
.
Willems
P.
,
Arnbjerg-Nielsen
K.
,
Olsson
J.
&
Nguyen
V.
2012
Climate change impact assessment on urban rainfall extremes and urban drainage: Methods and shortcomings
.
Atmospheric Research
103
,
106
118
.
https://doi.org/10.1016/j.atmosres.2011.04.003
.
Woltemade
C. J.
,
Hawkins
T. W.
,
Jantz
C.
&
Drzyzga
S.
2020
Impact of changing climate and land cover on flood magnitudes in the Delaware River Basin, USA
.
Journal of the American Water Resources Association
56
(
3
),
507
527
.
https://doi.org/10.1111/1752-1688.12835
.
Xu
H.
2006
Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery
.
International Journal of Remote Sensing
27
(
14
),
3025
3033
.
https://doi.org/10.1080/01431160600589179
.
Zha
Y.
,
Gao
J.
&
Ni
S.
2003
Use of normalized difference built-up index in automatically mapping urban areas from TM imagery
.
International Journal of Remote Sensing
24
(
3
),
583
594
.
https://doi.org/10.1080/01431160304987
.
Zhang
K.
,
Shalehy
M. H.
,
Ezaz
G. T.
,
Chakraborty
A.
,
Mohib
K. M.
&
Liu
L.
2022
An integrated flood risk assessment approach based on coupled hydrological-hydraulic modeling and bottom-up hazard vulnerability analysis
.
Environmental Modelling & Software
148
,
105279
.
https://doi.org/10.1016/j.envsoft.2021.105279
.
Zope
P.
,
Eldho
T.
&
Jothiprakash
V.
2016
Impacts of land use–land cover change and urbanization on flooding: A case study of Oshiwara River Basin in Mumbai, India
.
CATENA
145
,
142
154
.
https://doi.org/10.1016/j.catena.2016.06.009
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).