Sie sind auf Seite 1von 38

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Remote Sensing and GIS


Applications in
Forestry and Ecology
Dr. Arijit Roy
Scientist/Engr. - SF
Forestry and Ecology Department
arijitroy@iirs.gov.in

11/6/2015

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Our Environment
Earth provides enough to satisfy every man's need, but not every man's greed." M. K. Gandhi

unique, wonderful, enchanting


fragile, exploited, neglected

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Our Environment is Changing Rapidly


Becoming Increasing Challenged

Population density

Influenced by everything we do
and also dont do

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Driven by the Dynamics of


Increasing
Population
Resource Consumption
Development and Globalization
Political and Social Controversy
Technological Change

Threatening
Our Climate
Our Biodiversity
Our Natural Places
Our Stability and Security
Our Health

affecting the possibility of our future


sustainability of our environment

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

INDIAN FORESTS - SIGNIFICANCE


Indian forest ecosystems extend valuable
goods and services

Biodiversity is unique and has potential of


sustaining ecosystem functions and services

Labyrinth of biotic and abiotic factors


influencing the structure and functions need
state of the art technological assistance for a
holistic management
Sustainable use of forest resources requires
precise assessment of stocks of various
timber and non timber products

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

SPATIAL APPROACH IN FOREST RESOURCE


MANAGEMENT
Greenness

Crown closure

Climate
Topography
Socioeconomics

Vegetation type

Species assemblage/
Gregarious formations

Ground data
Volume
Biomass
Wildlife
Species Diversity

Water
Soil
Geology

I S

Multithematic, spatial, temporal Data


Multicriteria/Rulebased/Empirical/Mechanistic modeling
Forest cover

Forest type

monitoring

mapping

Growing
stock
assessment

Carbon
sequestration

Protected

areas

Forest
protection/
degradation

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Potential Remote sensing resolutions for Plant resource


assessment and conservation
Operational

Research and Development

Coarse pixel
Moderate
resolution

60-500 M

~ 30 M

Biogeochemical
interactions

High
resolution
~ 1 M to 5 M

Hyperspectral
Laser borne

220 bands

1ret/sq.m

Habitat
Condition/health

Ecosystem

Forest type

Scales addressed

Habitat/Fringe
structure

Fine scale
structure

Species
discrimination

Community
boundaries

Automated
Virutal 3-D model

System Nutrient
Status

community

Canopy

Withincommunity

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Multi-resolution satellite datasets for


deriving information across the scales

QB-MS

LISSIII

AWiFS
Shown are the series of
images using various
sensors in multispectral and
panchromatic modes

QB-MS ZOOM

QB-MS Multispectral
data of Quickbird satellite

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Spatial analysis tools for assessment of plant


resources
Geographic Information system

Data
retrieval
Avaialability
Site features

Data
analysis
Field gene banks

Cost surfaces

Site selection

Cellular
automata

monitoring
Trials
Generation

Spatial
modeling

Clines
Fragmented
Habitats
Exploration
scheme

Process
model
Habitat
resilience
Climate

Grazing,lopping

hydrology

Geostatistical
modeling

nutrients

phenotype
Pest/disease

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

To meet these challenges


We Need To Have A Common Vision:
Building a Digital Abstraction of the Earth

We Need Better Ways To

Represent

Understand

Manage

Communicate
.Our World As a System

An Evolving System for

Measurement
Monitoring
Modeling
Planning
Decision Making
Management

. . . Affecting Our Planets Evolution


Geoinformatics provides the framework

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Geoinformatics: A Scientific Technology


Relationships

Patterns

F
F
F

F
F

F
Fr

SF
Fr

Geoinformatics is a macroscope
for studying the Earth.

Processes

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

National Forest cover mapping


25
20

Hot Spots
Sensitive Areas

percentage

15
10
5
0

19721975
14.12
Dense Forest
Total Forest 21.6

A
A

19811985
10.88
19.52

19851987
11.51
19.47

19871989
11.71
19.44

19891991
11.72
19.47

19931995
11.73
19.45

19951997
11.17
19.27

19971999
11.49
19.39

19992001
12.68
20.55

20012003
11.88
20.64

Source NRSA

Source FSI

Forest Survey of India (Ministry of Environment and


Forest) has carried out six cycle of biennial vegetation
cover mapping. This has helped the country to identify

the districts/regions which are under rapid change of


forest cover

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Approach for Life Zone Mapping

Mean Annual Temperature


(NATMO Maps: Lines,
Climatic Database: 190 IMD Points,
Orographic correction applied
using elevation)

Annual Precipitation
(NATMO Maps: Lines,
Climatic Database: 190 IMD Points)

Potential Evapotranspiration Ratio


(Calculated from Orographically corrected
Temperature using Thornthwaite Method)

CEIL{[ln(PETR) ln(0.0625)]/0.006931} +
CEIL{[ln(RF) ln(3.125)]/0.06931} +
CEIL{[ln(TEMP) ln(0.375)]/0.6931} -110
Where,
ln Natural logarithm
CEIL Rounding function

Digital Elevation Model


(Digital Chart of the World, used for orographic
correction and altitudinal zonation of vegetation)

Life Zone Map


(Holdridge Life Zone Classification)

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Biome Level Characterization of Indian Vegetation


Application of IRS 1C WiFS data
Biome Map of India
(Derived from Vegetation Cover Type and Holdridge Life Zone)
Legend
200

200

400 km

SCALE

Legend (Life Zone)


Rain Forest (Lower Montane)
Rain Forest (Plains)
Wet Forest (Lower Montane)
Very Dry Tundra (Alvar)
Moist Tundra (Alpine)
Moist Forest (Sub Alpine)
Moist Forest (Montane)
Moist Forest (Lower Montane)
Moist Forest (Plains)
Dry Tundra (Alpine)
Dry Scrub (Montane)
Dry Forest (Montane)
Dry Forest (Lower Montane)
Dry Forest (Plains)
Thorn Woodland (Lower Montane)
Very Dry Forest (Plains)
Thorn Woodland (Plains)
Desert Scrub (Plains)
Desert (Plains)

Forest Cover
Temperate Conifer Forest
Temperate Broad Leaved Forest
Temperate Mixed Forest
Tropical Evergreen Forest
Tropical Semi evergreen Forest
Tropical Moist Deciduous Forest
Tropical Dry Deciduous Forest
Tropical Thorn Forest
Sub Tropical Forest
Conifer Forest
Sub Alpine Forest
Mangrove Forest
Degraded Forest
Abandoned Jhum
Land Cover
Alpine Meadows
Grasslands
Scrub/Shrub
Current Jhum
Coral Reefs
Plantations
Orchards
Valley Agriculture
Irrigated/Rainfed Agriculture
Fallow Land
Barren Land
Sand Dunes
Mud Flats
Swamps
Wet Pans
Dry Pans
Water Body
Settlement
Other Classes
Snow
Cloud
Shadow

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Biodiversity Characterization at Landscape Level: Methodology


Vegetation Type Classification

Phase I

Phase II
3750

4000

550

Regions

1260

400
Cold
Desert

500

1131

Indo-Gangetic
Plains

120 Vegetation
types/habitats mapped

1000

1696

Arid Zones

1500

1763

Deccan Plateau

16518 field plot data of


7606 species database
integrated

2000

2425

Central India

2500

16518
Eastern Ghats and East Coast

Hybrid Classification
approaches followed

2900

A&N
Islands

3000

3739

Total Number
of Sample Points :

Western Ghats

3500

Western Himalaya

Two Season IRS LISS III


used

North Eastern Region

Phase III

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Biodiversity Characterization at Landscape Level - Outputs

Fig 4a- Online geo-visualization and point query in BIS

Fig. 4b- Grid wise data sharing and dissemination in IBIN

High Fragmentation in the Western Himalayas and Western Ghats


North east India has less fragmentation and Disturbance
High Biological richness in North east India and Western Ghats
Dissemination through Web

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Biodiversity Information System

The motive of conceptualization of


Biodiversity Information System (BIS) is
collection and organization of the
available but distributed spatial and non
spatial database, into an interactive
system which is capable of presenting a
user friendly and easily accessible interface
to its clients;

A Centralized repository of BCLL Phase I,


II& III is available with interactive GUI for
data access and dissemination.

Biodiversity information system (BIS) enables


Interacting ,querying and region specific conservation planning
Overlay of spatial products with administrative boundaries and base details
Access and analyse 6000 strong point phytosociology data against vegetation type and
ecoregion
Access to ecological categorisation ( rare, endanagered ..) of species studies

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Forest canopy density estimation


Ocular Estimation in the Field

90 %

70 %

60 %

50 %

Digital Photo Image Segmentation Estimation in the Lab

88 %

70 %

63 %

55 %

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Forest type and density stratification using satellite


images
L
Very High Density Forest (>80%)
High Density Forest (60-80%)
VH

Medium Density Forest (40-60%)


Low Density Forest ( 20-40%)

VH

Teak Plantation
Scrub

Forest Blank
Creeks

Pan Image of a part of Andaman Islands showing different Forest


canopy density classes

Mudflat
Agriculture

Forest Canopy Density Map derived from Pan Image

(Bands 4,3,2)

(Bands 4,3,4)

'Forest Cover (fc) refers to the percentage of forest area within a polygon.

a) fc >80%
b) fc~ 80% - 60% fc
c) fc~ 40 - 60%
d) fc 40% - 20%
e) Fc< 20%

Dense forest
Continuous forest cover
Fragmented forest cover
Mosaic
Non-forest

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Different Forest Canopy Density Mapping Techniques


91 %

Legend
> 80 %
60 - 80 %
40 - 60 %
20 - 40
< 20 %
Non-Forest

Landsat TM satellite image of the test site


87%

Forest canopy density stratification using FCD Model


81%

Legend

Legend
> 80 %
60 - 80 %
40 - 60 %
20 - 40
< 20 %
Non-Forest
River Bed

FCD stratification through


visual image interpretation

> 80 %
60 - 80 %
40 - 60 %
20 - 40
< 20 %
NonForest
River Bed

FCD stratification through


object oriented image analysis

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Predictive modelling of LULC


MODEL STRUCTURE

SPATIAL PATTERN OF
LAND USE CLASSES
& DRIVING FORCES

LOGIT
REGRESSION

WEIGHTAGES OF
DRIVING FORCES

CHANGE IN
DRIVING FORCES

DATABASE
ORGANIZATION

PROBABILITY SURFACE
FOR ALL LAND USE
TYPES

CORRELATION

-LOGIT
REGRESSION
-TRANSITIONAL
PROBABILITY

ALLOCATION OF
CHANGES
(T-2)

CONDITIONAL
IMAGE
COMPOSITION

ACTUAL
LAND USE
(T-1)

RS/GIS DATABASE

LULC Conversion matrix of Goa


LULC
Class

LULC Class
WL

CL

FL

WB

BU

Forest

WL

+++
+++

+++++

+++

++

++++

++++

CL

++++++

++++

+++

+++++

FL

+++

++++

++++++

++

+++++

WB

++++++

+++++

BU

++++

++++++

Forest

+++++

++++++

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Impacts of climate change

Projected Changes in Annual Temperatures for the 2050s

Spatial distribution of linear


trends in estimated NPP from
1982 to 1999.
NPP was calculated with
mean FPAR and LAI derived
from GIMMS and PAL data
sets.

Projected Changes in Annual Precipitation for the 2050s

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Species Niche modelling for assessing climate change


impacts
Species
Distribution

Approach

Environmental
Variables

Model
Training

Environmental Variables

Niche
Model

IPCC Scenarios
2050 (A1)
Projected LULC
Projected Species range in 2050

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Climatic variables for


model input
1. Annual mean temperature

10

11

12

13

14

15

16

17

18

19

2. Mean diurnal range


3. Isothermality
4. Temperature seasonality
5. Maximum temp. of warmest month
6. minimum temp. of coldest month

7. Annual temperature range


8. Mean temperature of wettest quarter
9. Mean temperature of driest quarter
10.Mean temperature of warmest quarter
11.Mean temperature of coldest quarter

12.Annual precipitation
13.Precipitation of wettest month
14.Precipitation of driest month
15.Precipitation seasonality
16.Precipitation of wettest quarter

17.Precipitation of driest quarter


18.Precipitation of warmest quarter

iirs

19.Precipitation of the coldest quarter

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Topographic and Landscape Variables

LULC 1977

Elevation
Aspect
Soil Depth
Soil Texture
Slope

LULC 2012

iirs

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Potential niche of Agapetis serpens in different climate change scenario


in Darjeeling District

1977

IPCC 2050 a2

2012

IPCC 2050 b2

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Key Observations
The species range is shifting from
subtropical broadleaf to
temperate broadleaf for
probability range between 0.9-1.0
The range although is relatively
constant during 1977-2012 is
projected to increase in 2050
There is a probability that this
species will shift to temperate
conifer where this species is
generally not found.

iirs

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Impacts of Degradation
FCC of MSS image, January 1973

55% GLOBAL FORESTS ARE LOST TO SHIFTING CULTIVATION


IN INDIA 2.5 M HA ARE LOST NE STATES , E. GHATS

FCC of IRS P6 LISS III image, January 2004

The regeneration potential of the


vegetation is adversely affected as the
seedling and the seeds are destroyed.
Clear felling of sal in Orissa for commercial
and domestic timber and land
requirements

Around Rs.440 crores loss in terms of


seedling and regeneration loss

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Forest degradation in Amazon


1970

1973

1978

1976

1985

1991

1988

1996

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Forest fire management suite of resolutions


SWIR NIR RED

NIR RED GREEN

Active fire

IRS P6 AWiFS of 14March 2005

MODIS-Terra of 14March 2005

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Identification of swamp deer potential habitat in Jhilmil Jheel


Conservation Reserve, Uttarakhand, India using multi-criteria analysis
Identification of potential habitat.
Only 5.44% (2.06 km2) of the area of the Conservation Reserve was found to be highly
suitable habitat.
An area of 2.57 km2 (1.37%) outside the Conservation Reserve is considered to be highly
suitable habitat
IRS P6 LISS -IV Imagery (26
Feb. 2005)

Vegetation/Land Cover
Type Map

Potential Habitat
Map

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Establishment of Ecosystem Network and Wildlife Corridors

Network analysis in
geospatial domain

Graph Theory,
Circuit Theory,
Least cost path
analysis

Areas of Application
Species migration
Local species extinction
Ecosystem energy and
nutrient flows
Ecosystem losses

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Emerging Remote Sensing Tools

IKONOS image of Part of the Dehradun


valley

Avenue tree delineation using image


segmentation

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Emerging Remote Sensing Tools

Pine

Sal

Terminalia

Bamboo

Teak

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Emerging Remote Sensing Tools

SAR derived Plant Biomass Map

Field data 3D
perspective view-Lidar
applications
Multispectral ------------------------------------------Hyperspectral

AVIRIS Hyper spectral image Data


Cube Composed of Individual Images
Recorded at 224 Spectral Bands
Between 0.4 to 2.4 m, (Each at about
10 nm spectral Bandwidth).

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Conservation of Critical Ecosystems Using Space Based Information

Through web based


information system
ISRO

Space Based
Information
Other Research and
academic
organizations

Integrated analysis
of existing data and
generation of
required dataset in
matrix mode with
ISRO
Ground based Information

Identification of
biologically rich areas

Areas of research and data use


Conservation of biodiversity
Designing ecological and wildlife corridors
Sustainable bioprospecting
Baseline database for ecological studies
Fragmentation

Connectivity

Identification
of
Biodiversity
hotspots
&Hotspots of
change: from
global to
regional

Identification of regions for


conservation

iirs

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Changes in ecosystem structure and functioning Due to Climate Change


Time series MODIS-NDVI over
Himalayan region (2000-2014)

Parameters to be studied
over long term

Phenology
Productivity
Decomposition
Nutrient cycling
Biodiversity
Invasive species
Fire proneness

Modelled shift in treeline species


(Abies pindrow) in Western Himalayas
due to proposed climate change (IPCC
AR 5 2050 scenario)
Elevation shift in Abies pindrow under climate
change

Elevation shift in Taxus wallichiana under


climate change
>3500
3000-3500

3000-3500

ELEVATION

ELEVATION

>3500

2500-3000
2000-2500

2500-3000
2000-2500

1500-2000

1500-2000

1000-1500

1000-1500

500000

1000000 1500000 2000000 2500000 3000000


PIXEL COUNT
future

present

Geospatial analysis
0

500000

1000000
1500000
PIXEL COUNT

future

2000000

2500000

present

INFORMATION DRIVEN
SCALABLE
MULTI-DIMENSIONAL
MODEL BASED
PREDICTIVE

Multi-Institutional
Collaboration

Establishment of long term observatories


(LTERS) with multi-institutional partnership

iirs

I N D I A N I N S T I T U T E O F R E M O T E S E N S I N G, D E H R A D U N

Wisdom
+ significance

Knowledge
+ understanding

Information
+ context
How can we add value to data
and climb the mountain?

Data
Contact Details of the Faculty:
Email- arijitroy@iirs.gov.in
Tel- 0135-2524177

Das könnte Ihnen auch gefallen