A.K.M. Saiful Islam Associate Professor, IWFM, BUET December 2010 Email: [email protected].

Analysis of Planform Changes using Satellite

Images

A.K.M. Saiful Islam Associate Professor, IWFM, BUET

December 2010 Email: [email protected]

Changes of River Banks

Change detection using Satellite Images◦ Types and sources of Satellite Images◦ Image Acquisition Processes◦ Image Analysis Tools: ILWIS, ERDAS◦ ILWIS: A Open Source Software

Google Earth: ◦ As a free image source & analysis tool

Hands on Exercise using ILWIS and LandSat Images

Presentation Outline

Most of the Major rivers of Bangladesh are either Meandering or Braided.

Every year, millions of people are affected by erosion that destroys standing crops, farmland and homestead land.

It is estimated that about 5% of the total Floodplain of Bangladesh is directly affected by erosion.

Change of Bank Line

Jamuna Ganges Padma Upper Megh

na

Lower Meghn

aAverage width (m)

1984 9,720 4,367 5,689 3,406 6,661

1993 *11,220 4,693 7,116 3,391 8,897

Rate of change of width (m/yr)

  184 36 159 2 248

Changes during 1984-93

Jamun

a GangesPadm

a

Upper Megh

na

Lower Megh

na

Bank erosion rate (m/yr)

Left   *100 -20 38 7 66

Right *84 56 121 -9 182

Maximum bank erosion rate (m/yr) *784 665 620 NA 824Bank erosion (ha/yr) *5,020 2,240 1,800 48 1,172

Bank accretion (ha/yr) *890 1,010 233 49 402

Change of width of the rivers

Source: ISPAN, 1995 Note: * Rates derived for the period 1984-92

Bank Erosion/Acceration

Severity of ErosionTeesta River at Sundarganj

Satellite data is useful for monitoring changes of Bank lines.

However, careful inspection and Ground Truthing is necessary to verify the changes.

Using images of various Bands, False Color Composite images are developed to identify various features of the landscape.

Application of Remote Sensing for Monitoring Changes of Rive Banks

LANDSAT 5 TM of the Korotoa river

True Color Composite False Color Composite image

Spectral Images can be used

Low resolution (<=1km): ◦ NOAA, MODIS

Moderate resolution (<100m): ◦ Landsat, SPOT

High resolution (<5m): ◦ ICOKON, Quickbird, IRS, World View-2

Satellites

Spectral Reflectance of Water, Vegetation, Soil and Rock

12 3 4 5 7TM

Medium Resolution Sattelite

Landsat-7

Orbit: 705 km, 98.20 inclination, sun-synchronous

Swath width: 185 km (FOV = 150)

Revisit time: 16 days

Spatial resolution: 15 m (PAN), 30 m (bands 1-5, 7), 60 m (band 6)

SPOT-1, 2, 3, 4

Orbit: 832 km, 98.70 inclination, sun-synchronous

Swath width: 60 km

Revisit time: 26 days

Spatial resolution: 10 m (PAN), 20 m (Multispectral)

High Resolution Satellites

SPOT-5

Orbit: 822 km, 98.70 inclination, sun-synchronous

Swath width: 60 km

Revisit time: 2-3 days

Spatial resolution: 5 m (PAN) 10 m (Multispectral)

IKONOS

Orbit: 681km, 98.20 inclination, sun-synchronous

Swath width: 11km

Revisit time: 1-3 days

Spatial resolution: 1m (PAN), 4 m (Multispectral)

QuickBird WorldView-1 WorldView-2

23 m 6.5 m 6.5 m

  QuickBird WorldView-1 WorldView-2

Panchromatic (B&W)

450 - 900 nm 400 - 900 nm 450 - 800 nm

Multispectral:

Coastal Blue     400 - 450 nm

Blue 450 - 520 nm   450 - 510 nm

Green 520 - 600 nm   510 - 580 nm

Yellow     585 - 625 nm

Red 630 - 690 nm   630 - 690 nm

Red Edge     705 - 745 nm

Near-IR 1 760 - 900 nm   770 - 895 nm

Near-IR 2     860 - 1040 nm

High Resolution Satellites

High Resolution images need to be purchased.

However, some of the moderate resolution images can be downloaded freely.

Landsat TM and ETM images can be downloaded from USGS Global Visualization Viewer.

MODIS Images can be downloaded from LP DAAC web site at https://lpdaac.usgs.gov/lpdaac/get_data/wist

Sources of Satellite Images

http://glovis.usgs.gov/

False color composite of Landsat TM image at Shaghata

11 November 1989 4 November 2005

Bank Shifting of bankline from base year (1989) to present year (2010) in meterLeft Bank Right Bank

At the bridge Location

-198.94 +240.8

1 km upstream

-157.16 +190.5

1 km down stream

-34.2 +67.35

Shift of Bankline

Planform changes of the Korotoa River at Shahjadpur

November 11, 1989 November 17, 2000

November 04, 2005 January 21, 2010

Change of Banklines

Google Earth as High Resolution Image Source

Bank lines using Google Earth

February 19, 2006February 27, 2004

Shifting of bank line from base year (2004) to present (m)

At the bridge Location

-1.9 16.2

1 km upstream -1.3 15.3

1 km down stream

-2.9 15.3

Shift of Bank lines

Hands on Exercise on Planform changes of River

using Satellite Images

A.K.M. Saiful IslamIWFM, BUET

Input Data: ◦ Landsat Images of Year 1989 and 2010 of Band 1-4◦ Bridge Location Text File (Lat, Long)

Software: ILWIS 3.7

Assignment: ◦ Develop False Color Composite Images ◦ Show the location of the Bridge Site◦ Determine the Bank lines during 1989 and 2010◦ Detect Changes of Banks

Outline

Tiles of Landsat TM+ over Bangladesh

Arial Khan River at Kalkini of Madaripur district

Import Bridge Location Text File “Bridge_Sundar_Teesta.txt”

It will imported as Table Create a Point Map from the Table Assign coordinate as “Lat-Lon”

Import Bridge Location

Longitude, Latitude90.33028611, 23.09426389

Use Import Wizard, Select Text files, Press Next Button

Importing Bridge Locations

Import as “Comma Delimited” ILWIS Table

Import from Text files

Skip 1 Lines as Header

Formatting as value domain

Change domain as “value” Change precision as “1e-006”

Set precision as E-06

Finish importing table “kalkini_bridge”

Importing Bridge Text file as ILWIS Table

Open Table, Column1 as longitude and Column 2 as Latitude

Creating Maps file from Bridge Table

Create Point Map from the Table by Clicking right mouse > Table to Point Map

X Column as Long and Y Column as Latitude Coordinate System as LATLON Output PointMap “Kalkini_Bridge” > Show

Create Point Maps from Table

Point Map of Kalkini Bridge

Shows Bridge locationAs point map

Band 1, 2, 3, 4, 5, 6, and 7 in folder 2010 and 1989

Import using via GDAL as TIF format

Import of Landsat Images of 1989 and 2010 in ILWIS

Band 1-7 as TIF images

Import Geo-Tif Images to ILWIS format

Import using TIF format and write file name

Importing TIF Image of Bands

Change image format as (TIF) and give a name

Displaying image of a single band

Import Band 1 to 7 using the same tools with and save file name as b1 to b7.

Make a Map List using Band 1 to 7.

Import Band 1 to 7 from TIF Images

Make map list using “>” and name it as “maps”

Creating Maplists from maps

Add all 7 layers to create a maplist “Maps”

Select Bands (7-4-2) for false color composite maps

Use display tool

Select band 7-4-2 as RGB

False color image during 2010

False color composite of landsat image

Add Bridge Layer over the Image

Use add layer button

Select bridge point map and add it

Insert Bridge location on false color composite maps

Change color to red

Location of the bridge in red color

Zoom over the Bridge Site

Use zoom tool

Zoom the map over bridge site

Creating Segment Maps of Bank

Create new segment For bank lines

Write segment names

Customize Snap tools and tolerance for onscreen diziting

Change tolerance Values to small value

Customize Using right Mouse button

Onscreen Drawing of Bank lines

Start digitizing using Insert tool and then save

Insert tool

Planform change during last 21 years at the Bridge site

Bank lines in 2010 Bank lines in 1989

Change of Planforms using Images and Field verification

Field visit to the bridge site

Changes over 21 years

Bank Shifting of bank line from base year (2000) to present year (2005) in meter

At the bridge Location

-332 +398.57

1 km upstream

-266.55 +389.108

1 km down stream

-121.33 +206.10

Measuring shifts at Bridge

Existing Bank Protection Works

Thank you