CONTENTS Introduction Deciding on the best system of floodplain - - PDF document

3/29/2010 1

• Dr. Mohd Ekhwan , T

Assoc Professor, School of Soc. Development & Environmental Studies, FSSK..

• Assoc. Research Fellow. Institute for Environment and Development (LESTARI),

Universiti Kebangsaan Malaysia

International Symposium on Harmonising Environmental Considerations With Sustainable Development Potential of River Basin. 24th‐26th March 2010. Prince Hotel & Residence Kuala Lumpur

CONTENTS

Introduction‐ Deciding on the best system of

floodplain land classification Aim and objective floodplain land classification, Aim and objective

Study area‐ Langat Flooplain. Methodology‐ TWINSPAN & Multivariate Land

Classification Technique.

Results and Discussion:

Land class 1 Land class 2 & 3 Land class 4

Conclusion

3/29/2010 2

Introduction‐ Deciding on the best system of floodplain land classification.

What

is river floodplain? Floodplains are areas adjacent to rivers and coasts susceptible to flooding adjacent to rivers and coasts susceptible to flooding during periods of heavy rain and high river flows.

Typical river cross section showing floodplain area Floodplain level Bankfull Q FLOODPLAIN MEASUREMENT

3/29/2010 3

Why we need floodplain land classification system?

For managing a complex interaction between attributes (Physical & human interaction)

Natural Environment Human Environment

ADAPTATION & MITIGATION FLOODPLAIN SYSTEM

LIMITATION IN MANAGING THE FLOODLAIN

Many land use‐related disciplines has numerous

Many land use related disciplines has numerous classifications.

Each land‐management agency has several systems. Each geographic area varies in the applicability and

use of different classifications.

Unit area method‐ Tenness Valley Authority

Fractional code – Malaysia District

• ffice

DRASTIC system‐North Carolina Department of Environment

3/29/2010 4

ADVANTAGE OF ADVANCE COMPUTER GENERATING

Help planners and land managers tool to Floodplain

Land Classification Model.

Reduced time‐consuming particularly in the stage of

field data collection.

MSDOS interface

DECORANA TWINSPAN

Centre for Ecology & Hydrology Wallingford, Oxfordshire ECAM, Newcastle Upon Tyne

MSDOS interface & Windows version

AIM & OBJECTIVES

To develop the application of multivariate methods

which provide a basis for field sampling.

To

describe landscape variation and assess the distribution of natural and man‐made resources.

To provide a series of land classes. These samples

enable predictions to be made about the composition

• f each land class and the whole catchment generally.

3/29/2010 5

LANGAT RIVER CATCHMENT

Ca= 1815 km2 Total stream length = 2530.46 km Size of floodplain area = 508 3 km2 Size of floodplain area = 508.3 km2 Drainage density = 1.39 Stream order = 5

Rainfall pattern

3/29/2010 6

Drainage density (1.39)

TWINSPAN & Multivariate Land Classification Technique

The classification of the floodplain is based on floodplain

areas mapped in 1995 by the Drainage and Irrigation D (DID) h b i f h f fl d Department (DID) on the basis of the extent of past floods.

The Langat floodplain was divided into 508 one‐kilometer

grid squared.

TWINSPAN (Two‐way INdicator SPecies ANalysis) was

used to produce the multivariate classification.

54 attributes representing: topography, geology & soil,

vegetation, hydrology & geomorphology, human artifacts.

Public domain: Window WinTwin (Version 3.2) at:

http://www.ceh.ac.uk/products/software/CEHSoftware‐ DECORANATWINSPAN.htm

3/29/2010 7

Data from

• ne in 508

km squares

54 map attributes. Data includes topographical features, geology and soil, vegetation, hydrology and human artifacts.

Characteristic s from 508 km

TWINSPAN : Indicator species analysis. Classification into groups of square.

CLASS 2 CLASS 3 CLASS 4

55 divisions of 42 Land classes identified by TWINSPAN (Thi

CLASS 1

Characterize the classes by their constant and selective attributes.

TWINSPAN (This study only analyse based on 4 land classes)

RESULTS: TWINSPAN OUTPUT

From the original 508 grid squares=

g 5 g q

a.

144 were classed as land class 1,

b.

243 as land class 2,

c.

104 as land class 3 and

d.

17 as land class 4.

In general, the distribution of the Langat floodplain can be

divided into 3 general land classes: d v ded to 3 ge e a a d c asses:

a.

upland areas (land class 1),

b.

intermediate or transitional between lowland and upland areas (land classes 2 and 3) and

c.

lowland areas (land class 4).

3/29/2010 8

MULTIVARIAT LAND CLASSIFICATION

N 0 4 km LEGENDS Land class 1 Land class 2 Land class 3 Land class 4

Distribution of the Langat Floodplain land classes

CLASS 1 CHARACTERISTICS

Upland map characteristics predominate in this class,

which is located on the upstream valleys of Langat River d S ih Ri and Semenyih River.

Mostly the entire of the Hulu Langat District located

within this land class.

The landscape is influenced by Titiwangsa Range, which

surrounded with 100‐200m and more 200m contours.

Topography is steeply sloping where it is dominated by

serdang series serdang series.

The present of this soils indicate that the areas are

vulnerable to surface erosion.

3/29/2010 9

CLASS 2 & 3 CHARACTERISTICS

This is intermediate between lowland and upland

classes. Thi t f l d l t i ll th j b

This type of land classes contains all the major urban

areas: Kajang, Bandar Baru Bangi, Cheras, Dengkil and Semenyih.

Geologically consists of Stratigraphy: Quaternary was

dominated in the land class 2, which is about 47.6 per cent from the total grid squares. I h l d l l i i i l il l

In the land class 3, cultivation mainly oil palm estates

are predominant. It covers about 19.3 per cent from the total grid squares.

CLASS 4 CHARACTERISTICS

This is the lowland land class of the Langat floodplain

and down stream of the Langat River. g

It is coastal, generally close to intertidal sand or mud

and is found particularly around Banting and Klang towns.

Topography is fairy flat although short steep slopes are

common in areas of Jenjarum and Bukit Cheeding.

Wetland forest, mainly mangrove (Bakau), local

species such as nipah and gelam are predominate.

3/29/2010 10

CONCLUSION

The Langat Floodplain multivariate land classification

id i d h li i ifi i hi h provides an integrated, holistic stratification on which minimum affective field sampling programs can be based.

More importantly, it can also be used as a unifying system

• f classification, providing a common framework for many

users, including environmental management, forest management, urban management and nature i conservation.

In conjunction with standardized methods of field survey,

it provides a flexible, multipurpose approach to assessment and monitoring of natural and man‐made resources.

Thank you

Special Thanks to:

Universiti Kebangsaan Malaysia, Wallingford Software (UK) & Environmental Centre and Management (ECAM), Department of Civil Engineering University

• f Newcastle Upon Tyne which provide funding, computer and software

facilities during the 1 week 2009 summer training on Floodplain Modeling g 9 g p g Workshop.