Multi-platforms of remote sensing for mapping the water quality

(1)

Multi-platforms of remote sensing

for mapping the water quality

Cheng-Chien Liu

Professor, Department of Earth Sciences

Director, Global Earth Observation and

Data Analysis Center (GEODAC)

National Cheng-Kung University

劉正千

地球科學系/教授

全球觀測與資料分析中心/主任

國立成功大學

(2)

Multi-platforms of remote sensing for mapping the water quality 27 April 2012 2

Introduction



Significance of monitoring the coastal/inland waters in Taiwan



Frequent earthquakes and typhoons + steep mountains

 the

highest erosion rate in the world (Dadson et al., Nature, 2003)



High loading of suspended sediment in coastal regions



Deterioration of water quality in reservoirs after typhoon



More than half of water reservoirs are eutrophicated

Multi-stage observations of water quality both in space and time

Sustainable Water

Environment requires

(3)

Introduction



Efforts and progresses we made in the past 8 years



New data



Formosat-2

multispecitral imagery



Airborne Intelligent Spectral Imaging System (

ISIS-1

) imagery



Shipborne

HyperSAS



In-water Intelligent Spectral Imaging System (

ISIS-2

) imagery



New platform



CropCam Unmanned Aerial Vehicle (

UAV

)



Eco-Mapper Autonomous Underwater Vehicle (

AUV

)



New algorithm/software



Genetic And Semi-Analytical algorithms (

GA-SA

)



Export Landslide & Shaded Area Delineation System (

ELSADS

)



Automatic Mission Planning & Image Processing System (

AMPIPS

)

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

Multi-stage observations of water quality both in space and time

We endeavor to develop

the technique of

(4)

Remote sensing of water color



International Ocean Color Coordinate Group's effort



Compile the global database



Test various algorithms

Constituents

IOPs

AOPs

COMs

RTM

Forward problem

Inverse problem

Introduction

MODIS

GA-SA

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary



















(

ˆ

;

)

(

ˆ

;

)

(

ˆ

;

)

(

ˆ

;

)

(

ˆ

)

;

ˆ

(





_

_

_

_

_

_

_

_

_

_

d

L

c

dr

dL

(5)

GA-SA approach

Ocean color

semianalytical

model

(SA)

Genetic algorithm

(GA)

Generating potential

solutions

Not satisfy

Evaluating the

difference

(objective function)

Satisfy

Satellite observed

sea surface

reflectance

Simulated sea

surface

reflectance

Optimum solution

Individual IOPs

Objective function

Root of mean square error

between simulated and

observed sea surface

reflectance

Decision variables

9 parameters

(individual IOP, SA

(6)

Space-borne sensor – MODIS



Advantages



Designed for remote sensing of ocean color



Well-calibrated data with atmospheric correction



Large spatial coverage



Free of charge



Data available in two days



Disadvantages



Poor spatial resolution (1km x 1km)



Low temporal resolution

 clouds



No atmospheric correction for very turbid waters

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(7)

Liu, C.-C.

*

_{, Chang, C.-H., Wen, C.-G., Huang, C.-H., Hung, J.-J. and Liu, J. T. (2009) Using satellite observations}

of ocean color to categorize the dispersal patterns of river-borne substances in the Kaoping River, Shelf and

Canyon System. Journal of Marine System. 76(4), pp. 496-510.

Julian day

20

40

60

80 120 140 160 180

220 240 260 280

320 340 360

0

100

200

300 P

re

ci

pi

tation (m

m

)

0

100

200

300

400

500

600

700

2/15 3/ 7/2 28 4/29 5/22 6/30 11/ 2 8 11/ 1 4 10/ 6 7/ 27 12/25 H a it an g T y p h oon M a ts a T y p h oon Sanvu Typho o n Ta li m Ty p h o o n Khanun Typho o n D a m rey T y ph oon L o ng w a ng Typhoon 7/9

(8)

Frontal

pattern

11/28/2005

SSW (QuikSCAT)

SST (MODIS-Aqua)

Chl-a (mg m

-3

₎

_a

g

(443) (m

-1

)

NAP (g m

-3

)

(9)

Typhoon

induced

short-term

bloom

7/27/2005

SSW (QuikSCAT)

SST (MODIS-Aqua)

Chl-a (mg m

-3

₎

_a

g

(443) (m

-1

)

NAP (g m

-3

)

(10)

North-westward

pattern

7/02/2005

SSW (QuikSCAT)

SST (MODIS-Aqua)

Chl-a (mg m

-3

₎

_a

g

(443) (m

-1

)

NAP (g m

-3

)

(11)

Space-borne sensor – Formosta-2



Advantages



Taiwan owned and

controlled satellite



High-spatial resolution

(Pan: 2m x 2m; MS: 8m x 8m)



Middle spatial coverage

(24km x 500km)



Data available in one day



Formosat-2 automatic image

processing system



Disadvantages



Poor spectral resolution

(3 bands with 80-100nm in visible range)



Low temporal resolution

 clouds



No atmospheric correction at all

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(12)

Formosat-2 Automatic Image

Processing system (F2-AIPS)



Formosat-2 Automatic Image Processing system (Liu 2006 IEEE TGRS)



Gerald to Level-1A, conversion, display, search and archive



Band-to-band coregistration (Liu et al. 2007 IJRS)



Automatic orthorectification (Liu and Chen 2009 Optics Express)



Multitemporal imagery geometrical registration (Liu 2006 IEEE TGRS)



Multitemporal imagery radiance normalization (Liu et al. 2009 RAA)



Vicarious calibration of RSI/Formosat-2 (Liu et al. 2010 IEEE TGRS)



Disaster assessment



South Asia earthquake and tsunami (Liu et al. 2007 IJRS)



California wildfire (Liu et al. 2009 IJWF)



Wilkins ice shelf disintegration (Scambos et al. 2009 EPSL)



Environment monitoring



Polar region (Liu et al. 2008 AS) (Liu et al. 2008 CRST)



Illegal quarry of mining gravels on the riverbed (Liu et al. 2008 RAA)



Landslides in the catchment of water reservoir (Liu et al. 2011 IJRS)



The water quality of reservoir (Chang et al. 2009 JEM )



Site surveillance at airport/harbor (Liu 2006 IEEE TGRS)

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(13)

(14)

自然災害快速應變

2004

南亞地震與海嘯



打破操控的極限( 53

0 ₎



2004 南亞地震與海嘯

Reprinted from Liu (2006 IEEE TGRS)

Introduction

MODIS

Formosat-2

Natural Disaster

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(15)

自然災害快速應變

2004

南亞地震與海嘯



在眾多商業運轉的高解析度衛星中，福衛二

號是拍攝面積最大的衛星

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(16)

自然災害快速應變

2004

南亞地震與海嘯



產製的第一幅災損評估影像地圖

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(17)

自然災害快速應變

2004

南亞地震與海嘯



透過網路免費提供給國際上需要的組織

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(18)

自然災害快速應變

2008

四川大地震



打破操控的極限( 53

0 )



2004 南亞地震與海嘯



打破接收的極限(10

0 )



2008 四川大地震

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(19)

2006/5/14

(20)

後期影像

(21)

2008/05/14

(22)

2006/5/14

(23)

後期影像

(24)

自然災害快速應變

2007

加州山林大火



打破操控的極限( 53

0 )



2004 南亞地震與海嘯



打破接收的極限(10

0 )



2008 四川大地震



與其他感測器協同取像



2007 加州山林大火

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(25)

Source:

http://earthobservatory.nasa.gov/NaturalHazards/natural_haz

ards_v2.php3?img_id=14577

(26)

自然災害快速應變

2007

加州山林大火

Liu, C.-C.*, Wu, A.-M., Yen, S.-Y. and Huang, C.-H. (2009) Rapid locating of fire points from Formosat-2 high-spatial-resolution imagery: example of the 2007 California wildfire. International Journal of Wildland Fire. 18(4), pp. 415–422. (SCI)

(27)

自然災害快速應變

2009

宏都拉斯強震



打破操控的極限( 53

0 )



2004 南亞地震與海嘯



打破接收的極限(10

0 )



2008 四川大地震



與其他感測器協同取像



2007 加州山林大火



在打破時間紀錄的情形下獲取關鍵影像



2009 宏都拉斯強震

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(28)

自然災害快速應變

2009

宏都拉斯強震

Formosat-2 Satellite Orbit on 2009.5.28. T0: UTC Time of Earthquake Occurrence;

T1: UTC Time of Command Upload; T2: UTC Time of Urgent Imaging.

(29)

(30)

自然災害快速應變

2008

四川大地震



打破操控的極限( 53

0 )



2004 南亞地震與海嘯



打破接收的極限(10

0 )



2008 四川大地震



與其他感測器協同取像



2007 加州山林大火



在打破時間紀錄的情形下獲取關鍵影像



2009 宏都拉斯強震



災區定點監控



2008 四川大地震

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(31)

(32)

(33)

(34)

(35)

(36)

(37)

(38)

(39)

(40)

(41)

(42)

自然災害快速應變

2009

莫拉克颱風



打破操控的極限( 53

0 )



2004 南亞地震與海嘯



打破接收的極限(10

0 )



2008 四川大地震



與其他感測器協同取像



2007 加州山林大火



在打破時間紀錄的情形下獲取關鍵影像



2009 宏都拉斯強震



災區定點監控



2008 四川大地震



通過同一軌道進行五條帶取像



2009 莫拉克颱風

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(43)

13 August 2009

(

莫拉克颱風後第 5 天)

(44)

(45)

(46)

高雄縣小林村 (2009/2/4)

(47)

高雄縣小林村 (2009/8/24)

(48)

國家太空中心福衛二號密集取像（8月13日）

台東縣太麻里溪上游包盛社附近的堰塞湖

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(49)

國家太空中心福衛二號密集取像（8月13日）

南投縣信義鄉神木村和社溪上游的堰塞湖

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(50)

國家太空中心福衛二號密集取像（8月13日）

高雄縣桃源鄉荖濃溪上游支流拉克斯溪

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(51)

Collect the ground truth using a handheld

spectroradiometer to derive an empirical algorithm

Introduction

MODIS

Formosat-2

Natural Disaster Water quality

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(52)

Tseng-Wen Reservoir

(譚子健 2007)

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(53)

02/07

03/31

05/10

06/17

07/31

08/25

10/19

10/20

11/09

12/04

12/12

01/26

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(54)

Monitoring reservoir water quality with

Formosat-2 high spatiotemporal imagery

2006/2

2006/5

2006/6

2006/7

2006/8

2006/10

2006/11

2006/12

Chang, C.-H., Liu, C.-C.

*

_{, Wen, C.-G., Chen, I-F., Tam, C.-K. and Huang, C.-H. (2009) Monitoring reservoir water}

(55)

Test platform – CropCam®



Specification

Wing span

2.44 m

Length

1.22 m

Weight

2.72 kg

Payload

400 g

Power

1 hp

Engine

Brushless electric motor

Cruise speed

50 - 80 km/hr

Batteries

Lithium Polymer

Flight duration

25 min

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(56)

AMPIPS + CropCam

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(57)

Autonomous underwater vehicle

ECO-Mapper

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(58)

Autonomous Underwater Vehicle

AUV being deployed from the shoreside of the Hutopi Reservoir

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(59)

Deployment of AUV in Shinshan

Reservoir, Keelung, Taiwan

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(60)

Visualizing and publishing the AUV

measurements on Google Earth

AUV path

Study area: Hutopi Reservoir, Tainan, Taiwan

Instantly share the AUV monitoring data online

Chlorophyll-a products

Water surface temperature

products

(61)

Summary



We’ve been endeavoring to develop the technique of

Multi-stage observations of water quality both in space and time



Formosat-2 imagery + F2-AIPS + ELSADS



Airborne ISIS-1 + GA-SA



CropCam UAV + AMPIPS



Shipborne HyperSAS



In-water ISIS-2 + GA-SSA



Eco-Mapper AUV + ?



All data and products can be shared through our Web-GIS

powered by NCKU Cloud



There are still some rooms for platform/processing to improve



More applications need to be explored



Any type of collaboration would be more than welcome

Introduction

MODIS

Formosat-2

ISIS-1

UAV

HyperSAS

ISIS-2

AUV

Summary

(62)

Thank you for your attention

報告完畢，敬請指教

[email protected]

Cheng-Chien Liu

Professor, Department of Earth Sciences

Director, Global Earth Observation and

Data Analysis Center (GEODAC)

National Cheng-Kung University

劉正千

地球科學系/教授

全球觀測與資料分析中心/主任

國立成功大學