My Research Paper 2

Floodplain Modelling and Mapping through Spatial Technique

https://scholar.google.co.in/scholar?q=Floodplain+Modelling+and+Mapping+through+Spatial+Technique&btnG=&hl=en&as_sdt=0%2C5

ABSTRACT: Flood hazard is the chance that a flood event of a certain magnitude occurs in a given area, rural or urban within a given period of time. The severity of flooding tends to increase with urbanization. It is, therefore prudent that such a natural hazard is addressed in a way to reduce the impact it causes on people and the environment. To achieve the flood river modelling and mapping incorporating spatial technology and the HEC-RAS hydraulic model. The geometric data needed for the modelling process were derived from the observed data. Urban growth within the river basin has the tendency to increase both the volume and the rate of runoff. Socio-economic impacts of floods depend on the area, duration and depth of inundation, population density, etc. To achieve the aim of flood mapping using spatial technology, the envisaged study incorporating hydraulic model was used. The peak flow of the hydrograph derived from hydrologic model was used as an input in hydraulic model for developing flood inundation map. The increase in population and urban sprawl has contributed to the change in land use as people are converting floodplains to industrial and residential use. Comparing the LU/LC map of 1992 with the map of 2013, coverage of settlement area was found to have increased by 14.30% to 32.54%. Hence, for a given amount of rainfall, greater flooding is realized. The flooded area was geometrically overlaid on the Google Earth Image to delineate the affected buildings and slums. The floodplain area gets reduced due to unplanned development and encroachment. Flood inundation mapping of any rural area will be vital for emergency action plans, and ecological studies. The outcomes can help increase rural flood resilience of agriculture systems, livelihoods and habitats; reduce the impact of flood disasters by fore planning.

  

Keywords: Flood Modelling, GIS, Flood Inundation mapping.

INTRODUCTION

Nearly 37 million hectares (nearly 1/8th of India’s geographical area) of fertile land are prone to floods at one time or another during the monsoon (Valdiya, 2004). Understanding and assessment of land use change impacts on the watershed hydrologic processes, is of great importance for the prediction and mitigation of flood hazards, and also for the planning, sustainable development and management of the watershed. The study area has contributed to the change in land use as people are converting floodplains to industrial and residential use. Urbanization in floodplain areas increases the risk of flooding due to increased peak discharge and volume, and decreased time to peak (Campana and Tucci,2001; Liu et al., 2004; Nirupama and Simonovic, 2007; Saghafian et al., 2008). Even though flood is a natural hydrological phenomenon, human modification and new developments on floodplain had accentuated the problem. Aim of this research is to produce floodplain map based on the highest flood level (2005) of the last decade. The development of the present flood model integrates GIS with the hydrologic model and hydraulic model. Flood events and water surface profiles over the length of the modeled stream can be produce by using hydraulic model. The peak flow of the hydrograph derived from hydrologic model was used as an input in hydraulic model for developing flood inundation map. The hydrologic model determines the runoff and produces hydrograph as output. These hydrographs are given as input to the hydraulic model and the flood levels are calculated.

THE STUDY AREA

PCMC, Pune, Maharashtra, located to the North-West of Pune. The latitude of study area ranges from ‘18°33'45” N to 18°43'15” N’ and its longitude is between ‘73°42'30” E to 73°56'0” E’. It is covered by Survey of India (SOI) 1:50,000 toposheet numbers 47 F/10, 47 F/14. The township is situated at a height of 530 m above the sea level. Three rivers Pavana, Mula and Indrayani flow through this area. The study region has a typical tropical climate, with three distinct seasons- summer, monsoon and winter. The Leeward location with reference to the Western Ghats has made the area’s climate moderate and pleasant. The mean daily maximum and the minimum temperature for the hottest month - May is around 40^oC and 23^oC respectively. For the coldest month of December the temperature ranges from 30^oC to 12^oC. The relative humidity ranges from 36% in March to 81% in August. Three fourth of the annual rainfall of 70 cm is received during June to September and average annual rainfall is 700-800 mm.

OBJECTIVE OF THE STUDY

The main aim of the research is conceptualize the flooding pattern and inundation based on the highest flood level (2005) of the last decade. In addition, this study was extended to produce land use/land cover maps to understand the change detection and increase in settlement under floodplain within the study area for the years 1992, 2013. Furthermore, the velocity grid of flood and the affected area under agriculture and settlement has been worked out deriving data from Google map, which gives an idea about damage. 

METHODOLOGY 
Floodplain analysis was done using ArcGIS, HEC-GeoRAS, HEC-HMS and HEC-RAS. Flood maps were generated by exporting the hydraulic model output results to ArcGIS, where they were processed to identify the flooded area. The methodology used in the reported work consists of the following three steps and has been depicted in Figure 2.

1. Pre-processing of geometric data for hydraulic model, using GIS environment;

2. Hydrologic analysis and Hydraulic analysis;

3. Post-processing of hydraulic model results and floodplain mapping, using GIS environment. Process diagram for using HEC-GeoRAS is shown in Figure 1.

PRE-PROCESSING

Set of tools and utilities provided with the ArcGIS computer package is utilized in the preparation of spatial data for the hydraulic analysis. HEC-GeoRAS was used to create a geometric import file for import into HEC-RAS. GeoRAS requires the use of a Digital Elevation Model (DEM) in the form of a Triangulated Irregular Network (TIN). Cartosat DEM of 30m spatial resolution has been used in this study. 
Terrain TIN

The TIN format was used to generate attribute data for the Hydraulic model files. The contour lines at an interval of 1 m shape file are used in TIN development. Elevation data is extracted from the Terrain TIN and it is also used to locate the floodplain. 

Stream centerline

The stream centerline was digitized, using the toposheet for reference, from upstream to downstream. The stream centerline is used to calculate the river stationing at each cross section. 

Bank lines

Bank lines were located at the edge of the main channel. The bank lines used with the cross-sectional cut lines to determine bank stations at each cross section. 

Flow path centerlines

Flow path lines are used to calculate the downstream reach lengths in the left overbank, main channel, and right overbank between cross sections. 

Cross-sectional cut lines

Cross-sectional cut lines are digitized from the left to right overbank (when facing downstream). Cut lines are created perpendicular to flow. 

HYDROLOGIC ANALYSIS

The model divides a watershed into subbasins. It converts an amount of rain into runoff at exist of the each subbasin, and routes this runoff along the reaches until the outlet of the watershed. Combining a basin model, meteorologic model, and control specifications formed a simulation run. The hydrologic model determines the runoff and produces hydrograph as output. These hydrographs are given as input to the hydraulic model and the flood levels are calculated. 

HYDRAULIC ANALYSIS

Several RAS themes were created in the process of developing the geometric data file for import to HEC-RAS. The peak flow of the hydrograph derived from HEC-HMS was used as an input in HEC-RAS for developing flood maps. Flood events and water surface profiles over the length of the modeled stream were produced by using hydraulic model. The peak flow of the hydrograph derived from hydrologic model was used as an input in hydraulic model for developing flood inundation map. Hydraulic calculations are performed at each cross section to compute water surface elevation, water depth, and velocities. With the companion GIS utility, HEC-GeoRAS, those water surface profiles converted to flood inundation map.

POST-PROCESSING

Results exported from HEC-RAS were imported into the GIS using GeoRAS and the floodplain was delineated. Data exported from HEC-RAS included water surface elevations at each cross section, velocity information at distributed points along each cross section, and bounding polygon information. The bounding polygon information defined the extent of each cross section as modeled in HEC-RAS for the given flow. Floodplain delineation and velocity data were developed which adhered to the bounding criterion. Water surface profile data and velocity data exported from HEC-RAS was processed for floodplain delineation, inundation depth, and velocity GIS data sets.

Bounding polygon

The bounding polygon information defined the extent of each cross section as modeled in HEC-RAS for the given flow. During the post processing of HEC- RAS results, GIS layers for inundation depth and floodplain boundary were created. The water surface TIN is clipped by a bounding polygon.

Water surface TIN

HEC-RAS is used to determine the water surface profile for the Mula and Pavana River. Cross sections, exported from HEC-RAS and tagged with water surface elevations, are used as break lines to develop the water surface TIN. The water surface TIN was generated by intersecting the water surface elevations with the cross section cutlines. Thus, the floodplain polygon was created by intersecting the water surface TIN with the terrain surface. The water surface TIN was generated by intersecting the water surface elevations with the cross section cutlines.

Floodplain delineation

A floodplain polygon is determined by vectorizing the boundary of the water depth grid. Floodplain delineation and velocity data were developed which adhered to the bounding criterion. Floodplain delineation was performed resulting in an inundation depth and floodplain polygon.

Water depth

With the companion GIS utility, HEC-GeoRAS, water surface profiles converted to the flood inundation map. The flood depth map is obtained by subtracting the TIN from the water surface. The depth of water induced by the flood varied in different parts. A higher depth of flood is associated with a high discharge, which is a determining factor in flood-induced destruction of life and property.

Velocity

The simulation produced variable flow velocities in the main channel and the inundated floodplain. Velocity grids were developed from the cross-sectional velocities. As the slope of river increased the velocity of water also increased therefore the high values at downstream is accredited to the steep slope of the terrain whiles the low velocities at the upstream can be attributed to the flatness of the terrain.

LAND USE/LAND COVER CLASSIFICATION

For the current research, two Landsat scenes of the study area acquired on 4 December 1992 and 14 December 2013 have been obtained and images were at a spatial resolution of 90m, 30m respectively. The land use/land cover maps were extracted from LANDSAT 5 TM, LANDSAT 8 OLI/TIRS satellite image and the data was classified using maximum likelihood supervised in ERDAS Imagine 9.2 software. The image was taken through four stages to generate a land cover classes of the study area. These include: image pre-processing; feature extraction; selection of training data (signatures); and selection of suitable classification approaches. A remotely sensed image was classified into various land cover types, which is showing increase in settlement area. The following five land cover and use classes were obtained; Agriculture, Fellow land, settlement, Vegetation and Water bodies. Following Comparison of the LU/LC map of 1992 with the map 2013 was done to understand the land use change in the study area and to identify the increased settlement along the river bank and under floodplain. Figure 5 and Figure 6 shows the classified image with five classes.

OVERLAYING ON GOOGLE EARTH

Flooded area was geometrically overlaid on the Google Earth Image to delineate the affected buildings and slums. Settlements along the river bank (2005) are shown in Figure 8 in which area marked with red color boundary indicates the affected area of the study region under floodplain of Mula River and Pavana River.

RESULTS AND DISCUSSION

Flooding pattern and inundation based on the highest flood level (2005) of the last decade is conceptualized. About 6.06% PCMC was under water in hazardous 2005 flood event. Total area of Pavana and mula drainage basin is 144.44 sq. mile and outlet discharge of 26 July 2005 was 41642.8 cusec. For 2005 flood, flood level was defined, which was 0 feet to 10.16 feet which correspond to the peak discharge of Pavana River and Mula River respectively 27727.9 cusec, 14499.6 cusec. Generally, high water depth occurred along the main channel and spreads gradually to the floodplains. The inundation map produced clearly shows the spatial distribution of the flooded area which is located at areas with relatively low relief. The model results gave a maximum velocity of 2.25m/s with high velocities occurring mostly in the main channel. Generally, high velocities were recorded in the main channel than the floodplains. The spatial distribution of inundation flow velocity of the catchment shows a correlation with the spatial distribution of the elevation as low values flow velocities are observed at upstream and high values at downstream. Comparing the LU/LC map of 1992 with the map of 2013, fellow land coverage was found to have decreased by 46.61% to 15.99% respectively, whereas the coverage of settlement area is found to have increased by 14.30% to 32.54%. By overlying floodplain polygon of area on Google Earth Image, settlements on the bank of River Pavana and River Mula are shown. These settlements are constructed in river floodplain, hence can be affected by a miniature flood event also. Because of increase in settlement along the river bank threat of flood hazard has increased due to damage related to human and property loss.

CONCLUSIONS

The study undertaken depicts the floodplain mapping of the area of interest for highest flood level 2005 which thrive inside of the extent of inundation in the area. This when overlaid with the Google map showed the areas under flood threat. Spatial distribution of flood water for a particular flood event can be well understood for a rural area thus incepted damage in impact analysis can be carried out. Thus conserving national property in life the settlements affected by any flood threat can identify the hazard indication to human and property and thus remedial measures and rescue operations can be plan well in advance. This technology will say the prospects in the promising future if implemented in areas of flood prone.

REFERENCES

1.       Campana, N.A., Tucci, E.M.C., 2001. Predicting floods from urban development scenarios: case study of the Diluvio basin, Porto Alegre, Brazil. Urban water 3, pp.113–124.

2.       Liu, Y.B., De Smedt, F., Hoffmann, F., Pfister, L., 2004. Assessing land use impacts on flood processes in complex terrain by using GIS and modeling approach. Environmental modeling and assessment 9, pp.227–235.

3.       Nirupama, N., Simonovic, S.P., 2007. Increase of flood risk due to urbanization: a Canadian example. Natural Hazards 40, pp.25–41.

4.       Saghafian, B., Farazjoo, Hassan, Bozorgy, Babak, Yazdandoost, Farhad., 2008. Flood intensification due to changes in land use. Water Resource Management 22, pp.1051–1067.

5.       Tate E., D.R. Maidment, F. Olivera and D.J. Anderson, P.E., 2002. Creating a Terrain Model for Floodplain Mapping. Journal of hydrology engineering, 7: pp.100-108.

6.       US Army Corps of Engineers. Hydrologic Engineering Center. 2008. HEC-RAS, River Analysis system, Version 4.0. Davis, California.

7.       US Army Corps of Engineers. Hydrologic Engineering Center. 2009. HEC-GeoRAS, Version 4.2. Davis, California.

8.       US Army Corps of Engineers. Hydrologic Engineering Center. 2010. HEC-HMS, Hydrologic modelling system, Version 3.5. Davis, California.

9.       Valdiya. K.S., Lessening., 2004. The ravages of floods. In book: Geology, Environment and society Universities Press,India, pp.112-115.