Application of RS GIS

1
Application of RS / GIS
Example 1 Coastal water quality
2

• Satellites provide a means for looking at a very
  large area of the world within a very short time
  period.
• Advantages
• Satellite imaging is desirable because it can
  cover relatively large areas (spanning several
  kilometers) at relatively low cost.
• Some limitations
• Spectral, radiometric, temporal resolution
• Cloud cover can block the reflected light,
• Water may be too deep or too turbid.
• Sun glare from the sea surface also can cause
  interference, resolution of some satellite images
  may be too low

3
Remote sensing in shallow coastal waters can
pose problems The water bodies are frequently
too small for satellite data to be useful.
Ocean color algorithms developed for open-ocean
satellites generally are problematic because of
the variety of sediments, submerged aquatic
vegetation, phytoplankton, and dissolved material
that affect the color of coastal waters and thus
the calculation of parameters within the water
column. Even if the algorithms are useful,
processing a raw satellite image into a
user-friendly geographic information system (GIS)
data product can require too much time,
equipment, and expertise for some management
purposes.
4

• GIS-BASED ANIMATION TOOL FOR MODELING STUDIES
• EFFECT OF URBANIZATION ON SUSTAINABILITY OF WATER
  RESOURCES
• HYDROLOGICAL SENSITIVITY OF COASTAL WATERSHEDS
• ASSESSMENT OF EXTENT OF FLOODING UNDER CLIMATE
  CHANGE CONDITION
• FLOOD ALERT AND CONTROL SYSTEMS
• MATTER-ELEMENT MODEL OF INTEGRATED RISK
  ASSESSMENT FOR FLOOD CONTROL SYSTEMS

5

• Identification of problems
• Monitoring objectives
• Project Management

6
(No Transcript)
7
Good weather
Collocated Spatio-temporal surveying
8
Image rectification
9
Data processing
Multiple Regression modelling
Applying model To image formula
Generation of Classes
10

• Insertion of multiple
• Regression formula
• Clustering
• Export to GIS format

Criteria lt2.0 ug/Loligotrophic 2.0-6.0 ug/L
mesotrophic 6.0-40.0 ug/l eutrophic gt40.0
hypertrophic
11

• Some other applications
• GIS-BASED ANIMATION TOOL FOR MODELING STUDIES
• EFFECT OF URBANIZATION ON SUSTAINABILITY OF WATER
  RESOURCES
• HYDROLOGICAL SENSITIVITY OF COASTAL WATERSHEDS
• ASSESSMENT OF EXTENT OF FLOODING UNDER CLIMATE
  CHANGE CONDITION
• FLOOD ALERT AND CONTROL SYSTEMS INTEGRATED RISK
  MANAGEMENT

12
Application of RS / GIS
Example 2 Numerical weather and ocean Prediction
13
Ocean and atmospheric model prediction Current
trends and issues Model development Data
assimilation Multi-model approaches Downscaling
applications Model nesting
14
Ocean models
Increased computer resources resolution of
basin-wide models has increased to the point
where it can be called eddy-permitting continued
development of physical ocean models has reached
a state to allow for basin-wide applications of
these concepts Systematic model intercomparison
studies have become possible.
15
Advantages of ocean and atmospheric
models Physically based models (compared to
statistical models) Predict 3D ocean/atmosphere/la
nd Coupled models provide the best long-term
potential for Skillful seasonal
forecasts Disadvantages Complex, Highly
technical Drift model error Need for extensive
Skill assessments
16
Importance of coupled ocean atmosphere models
Major basin scale oscillations, eg ENSO, result
from a coupled interaction between ocean and
atmosphere. The memory of the couple system
resides in the ocean subsurface thermal structure
where the atmosphere responds interactively to
the variations in the SST.
17
(No Transcript)
18
Coupled model Initialization Initial
conditions Ocean/Atmosphere/Land
surface Present schemes for ocean modelling are
relatively old compared to NWP (such as
univariate Optimal interpolation) Current
developments Advanced data assimilation (Mainly
subsurface and surface temperature and altimeter
data) More frequent observations Coupled
initialization schemes
19
Numerical atmospheric model
10-m wind stress, turbulent and radiation heat
fluxes
Sea surface temperature
Grid interpolations and conversion, units
conversions
Numerical oceanic model
20
Developments assisting numerical model
initialization
1980s Ship Sparse network temperature 1990s
Regular array in the tropical pacific
temperature available down to 400m
some surface winds data started to be available
Introduction of altimeter data global
sea level 2000 onwards Moving towards
availability of Global Temperature and
Salinity
21
Data assimilation
Imperfect Observations
Imperfect Forecast background
Data assimilation
Optimal estimate of unknown quantity uncertainty
22
The objectives of DA is to produce regular,
physically consistent 4-dimensional
representation of the atmosphere/ocean from a
heterogeneous array of in situ and remote sensing
instruments which sample (irregularly and
imperfectly) in space and time Consistency
comes from the use of models Meaning it
integrates theory (via models) with truthing (via
instruments) This process can also lead to
improvements in both models and instruments due
to the strong interaction
23
Data assimilation cycle
Model analysis
Observations
Forecasts Time 00
Model analysis
Observations
Forecasts Time 06
24
(No Transcript)
25
TMI-derived SST (in oC) on 04.08.99 _at_ 21.7UT.
This was used to nudge model variables.
Data assimilation
Final forecasted SST field (in oC) at 06.08.99
at 00UT. ?24 hrs
TMI-derived SST (in oC) at 05.08.99 _at_ 21UT.
26

• Ocean data assimilation critical for basin scale
  oscillations such as ENSO
• Very hard to test true real-time skill using
  long-hind-cast period an
• Alternative could be to use the most recent case
  studies.
• Skill scores
• Set of verification scores (NWP approach)
• Bias
• Indices such as Rms, anomaly correlations
• Deterministic Spatial anomaly correlation, mean
  square skill
• Multi-model versus single model

27
High-resolution numerical atmospheric model
COADS
TMI data
AVHRR data
Air-sea flux database
SOC atlas
High-resolution numerical oceanic model
28
Key priorities for the future Improve our
understanding of main modes of variability (e.g.
ENSO, NAO) Improve models convective
processes Use of advanced data assimilation to
exploit sparce ocean observing system Adapt a
coupled approach at the model initilisation
stage Build links with user community
(application models) and add value to forecasts