Software and Tools Overview Dream Ocean Satellite Image Workshop CH2M Hill Alumni Center, Corvallis, Oregon August 18-19, 2011 Ichio ASANUMA The Tokyo.

Software and Tools Overview

Dream Ocean Satellite Image WorkshopCH2M Hill Alumni Center, Corvallis, Oregon

August 18-19, 2011

Ichio ASANUMAThe Tokyo University of Information Sciences

Software and Tools Overview

• Tools and operating systems• Data levels and manipulation• Geophysical parameters• Spatial and temporal analysis• Models for spatial and temporal estimate• Models for decision making

Tools and operating systems

• Operating systems and PCs– Significant improvements of PCs as workstations,

desk-top or notebook computers

• Changes of windows system– Entertainments to business– Sometimes slowing our use by version change

• Stable dissemination of linux system– Commercial to freeware linux system

• Improvement of VMware player– Realization of two operating system within one

computer

Large disk space

VMware player• Implementation of linux with Vmware

player within windows system

Windows operating system

C drive forWindows

/ drive forLinux VMware

LINUX operating system with VMware player software

General image processing software IDL ENVI EARDAS MATLAB Geographic

InformationSystem

Radiance Geophysical data data

RegionalInstantaneous

GlobalTime seriesValue added

Mission orienteddata processingsystem IPOPP, SeaDAS

Sharing rolesin data levels and manipulation

• Mission oriented data processing system– IPOPP– SeaDAS

• General image processing software with geophysical parameters– IDL, ENVI, EARDAS, MATLAB, etc

• Geographic information system– ArcGIS

Mission oriented data processing system

• IPOPP• SeaDAS

• Future requirements– Continuity to the future missions– Share of roles with other processing or GIS

systems

General image processing software

• General image processing software works on geophysical parameters and with geolocation data

– Inputs: GeoTiff data & others– Processing: • Spatial composite and/or time series analysis• Modeling

– Outputs: GeoTiff data– Future requirements:• Cost

Example using general image processing software

• Asanuma, 2006, Depth and Time Resolved Primary Productivity Model Examined for Optical Properties of Water, Elsevier Oceanography Series 73, 91-109.

• Inputs to IDL under SeaDAS– Chlorophyll-a– Sea Surface Temperature– Photo-synthetically Available Radiation

• Outputs from IDL– Primary productivity

Application with general image processing software

10

SST Chl-a PAR

MODTRAN PAR (June)

PPeu=∫t∫z Pb(z,PAR(z,noon),T) C(z)

PARM(0,t)/PARM(0,noon)dz dt

Primary productivity modelAsanuma 2002

11

Vertical distribution of Chl-a (EdPAR)

-160

-140

-120

-100

-80

-60

-40

-20

0

0.01 0.10 1.00 10.00 100.00EdPAR (% to 0m)

Dep

th (

m)

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00Chl-a (mg/ m3)

0.05 0.10.20.30.81.57.5100.40.50.60.70.811.5251.50.10.20.30.40.50.60.70.8125

Vertical distribution of Chl-a and PAR

Vertical distribution of EdPAR for chlorophyll-a concentration for 5.0

mg/m3.

Vertical distribution of chlorophyll-a

concentration for 5.0 mg/m3.

Asanuma 2001

12

In-situ and satellite estimated primary productivity

0

500

1000

1500

2000

0 500 1000 1500 2000

In-situ PP (mgC.m-2.day-1)

Sat

ellit

e es

timat

ed P

P(m

gC.m

-2.d

ay-1

)

Sub- tropicSub-arcticEquatorial Pacific

Asanuma modelVersion 2002.11Correlation coefficient = 0.768 r2 = 0.590

Asanuma 2002Validation of the model

Example using general image processing software

• Zainuddin et al., 2006, Using multi-sensor satellite remote sensing and catch data to detect ocean hot spots for albacore (Thunnus alalunga) in the northwestern North Pacific, Deep-Sea Research Part II 53, 419-431.

• Inputs to IDL– AVISO mean sea-level anomaly (MSLA)– Chlorophyll-a– Primary productivity– Sea surface temperature

• Output from IDL– Probability of catch rate of albacore

Application with general image processing software

Albacore CPUE

Albacore/boat-days in 1999-11 on TRMM/TMI SST(contour on 20 deg-C)

Albacore/boat-days in 1999-11 on SeaWiFS chlorophyll-a (contour on 0.3 mg m-3)

Albacore/boat-days in 1999-11 on an environmental provability mapgenerated from SST and Chl.

Geographic information system (GIS)

• GIS approach– GIS controls layers and provides value added

maps– GIS could be useful in real world decision

making.

• Simple approach by GIS– Boolean logic with un-weighted layers

• Complex decision support by GIS–Multiple criteria, multiple objectives weighted

variable layers

GIS integration

• Global, time series and value addition– ArcGIS

• Inputs: GeoTiff data/Layer components• Processing:– Insert additional information– Projection etc.

• Outputs: GeoTiff data/Layers• Future requirements: >> Connecting effort between RS & GIS

Primitive use of GIS

GIS produces value added maps with

GIS components of points, lines, or

polygons in conjunction with raster data and

control capability of layers.

GIS integration

• Multi-criteria decision tools– ArcGIS

• Inputs: GeoTiff data/Layer components• Processing:–Multi-criteria/Multi-objective decision making– Value added data production

• Outputs: GeoTiff data/Layers• Future requirements:– Algorithm implementation to GIS software

Decision making with GIS

GIS decision approachMulti-criteria evaluation (MCE)

• Methodology– Determination of criteria (factors)– Normalization of factors / criterion scores– Determination of weights for each factors

• Analytical hierarchy (AHP) process to calculate weights

– Evaluation using MCE algorithms– Sensitivity analysis of results

• AHP & MCE are functions of ArcGIS

Application of GIS multi-criteria evaluation

• Radiarta & Saitoh, 2009, Biophysical models for Japanese scallop, Mizuhopecten yessoensis, aquaculture site selection in Funka Bay, Hokkaido, Japan, using remotely sensed data and geographic information system, Aquacult. Int. 17:403-419.

• Inputs to ArcGIS– Sea surface temperature– Chlorophyll-a– Suspended solid bathymetry– Scoring and weighting parameter

• Output from ArcGIS– Aerial distribution of suitable aquaculture site

Why RS and GIS?(from JPL-ESRI report, 02, 2011)

• Characterize and understand complex process using measurements from multiple sources.

• GIS benefits:– To visualize, analyze, and overlay geo-

referenced data– To access to the actual data values– To access to a suite of robust analytic tools.

What are the problems that require connecting RS and GIS?

(from JPL-ESRI report 02, 2011) • Basic questions:– Where is data? – How much does data cost?– What data is necessary?– What knowledge is necessary to have for it?– What HW/SW is necessary?– How can we share the way with clients/users?

• For data access:– How can I gain access and how can I share?

• For analysis:– How do I go from data to information?

Key gaps or barriers to the use of RS data by GIS application

(from JPL-ESRI report 02, 2011)

• Many RS users do not need GIS to accomplish their work.

• Difficulties in integrating raster and vector dataset in GIS.

• Large volume of RS dataset for GIS.• Difficulties in integrating GIS software

with other applications.• Difficulties to deal multi-dimension

dataset in GIS.

http://support.esri.com/en/downloads/datamodel/detail/21

Summary• Hardware and operating system provide

more possibility in further application of satellite data.

• Mission oriented data processing system is the important function to support geophysical data to end user.

• General image processing software opens a possibility to implement new approaches by end users working on geophysical data.

• GIS provides capabilities to generate value added map and decision making tools.