area: monitoring based on repeated ground ... tools for wetland monitoring. 4. Remote sensing .... Application for monitoring: reedbeds evolution. 21. Influence of ...
USE OF REMOTE SENSING FOR MONITORING WETLAND PARAMETERS RELEVANT TO BIRD CONSERVATION AURELIE DAVRANCHE
TOUR DU VALAT ONCFS UNIVERSITY OF PROVENCE – AIX-MARSEILLE 1 UFR « Sciences géographiques et de l’aménagement » University - CNRS 6012 E.S.P.A.C.E
Camargue : Rhône river delta Dynamic system: water and sediment inputs from the Rhône and the sea
90 000 ha of natural habitats mostly wetlands 2/3 on relatively small private estates 2
Socio-economic activities and natural habitats Rice growing
Reed harvesting
Waterfowl hunting
Cattle grazing
Water management input of freshwater in brackish marshes modification of the hydroperiod division of the marshes into smaller dyked units
Influence on floristic composition and vegetation biomass Changes in bird habitat 3
Main objective Global loss of biodiversity
Proliferation of invasive species
Necessity to monitore the management and the health state of these marshes Reserve managers and stakeholders are in needs of management advices
A fragmented configuration within a large geographical area: monitoring based on repeated ground measures difficult
Remote sensing: good potentialities for wetlands spatial analysis
Development of reliable and replicable remote sensing tools for wetland monitoring
4
Specific objectives These tools will help to : ►map the vegetation of Camargue marshes (common reed, clubrush, aquatic beds) to follow their spatial evolution over time
►map flooded areas irrespective of vegetation density to follow their spatial evolution monthly ►map vegetation parameters that are associated with ecological requirements of vulnerable birds in reed marshes 5
Methodology Sampling
Image acquisition
Image processing
GPS
Vegetation characterisation (reedbeds, clubrush, aquatic beds)
Estimation of water levels for each image
Data image extraction Database Multispectral and multitemporal index
Statistical modellings: Classification trees Generalized Linear Models Formulas = maps 6
Sampling Fields campaigns : reedbeds, club-rush, aquatic beds, water levels, GPS
Digitalizations : Others 7
Image processing: radiometric normalization 6S atmospheric model vs. pseudo-invariant features (PIF) Similarity index (Euclidian distance): Estimation of radiometric variation of PIF 0.16
12
0.12
8
0.08
4
0.04 0
Roof
16
Each PIF varies at least once over the year
0
Sand
0.16
12
0.12
8
0.08
4
0.04 0
Necessity of different types of PIF
0
Dec
Mar
Radiometric variation (%)
Radiometric variation (%)
Pine tree
Water
16
May
Jun
Jul
Sep
Dec
Mar
May
Jun
Jul
Sep
6S does not take into account this variation for the correction
6 5 4
Variation significatively lower with 6S
3 2 1 0
6S
PI
8
Spectral variations 0,3
Reedbeds Club-rush
0,25
Aquatic beds
Influence of : • phenology
0,15
• pluviometry • water management
0,1
0,05
December
March
May
June
July
MIR
B3
B2
B1
MIR
B3
B2
B1
MIR
B3
B2
B1
MIR
B3
B2
B1
MIR
B3
B2
B1
MIR
B3
B2
0 B1
Reflectance
0,2
September
Natural and artificial phenomena characterizing Camargue wetlands
require
a
multispectral
and
multitemporal
imagery for their monitoring 9
Statistical modelling : two approaches 1 - Qualitative approach : presence/absence • Presence of reed, club-rush and aquatic beds • Presence of water in differing conditions of vegetation density Classification trees
2 - Quantitative approach : prediction of continuous variables • Diagnostic parameters of reedbeds • Quality for reed harvesting • Suitability for vulnerable reed birds species (passerines, Purple heron, Eurasian bitterns) Generalized Linear Models 10
Classification tree algorithm Rpart based on the algorithm CART (classification and regression tree) Breiman et al, 1984; implemented in R.
Method Recursive partioning based on gini index
Binary tree
Advantages Hierarchical classification strategy: easy interpretation of results
Optimal for presence/absence
Cross-validation (k-fold)
Small samples and reproducibility
Prior parameter
Unbalanced samples 11
Recursive partioning A two-dimension example with two variables selected for reedbeds classification 0,7
Split at 0.04897
0,6
Split at 0.2467
0,5
0,4
osavi12
0,3
other aquatic beds
0,2
reedbeds club-rush
0,1
0 -0,2
-0,15
-0,1
-0,05
0
0,05
0,1
0,15
0,2
0,25
-0,1
-0,2
-0,3
c30603
12
Tree: example for reedbeds classification c30603< 0.04897 2| 672/46
osavi12>=0.2467 2 128/46
1 544/0
ndwif209>=-0.3834 2 48/46
1 80/0
1 39/0
2 9/46
Reedbeds
Formula Presence of reedbeds = c30603≥0.04897 & OSAVI12=-0.5092 1 8/33
1 21/12
1 8/22
2 0/11
Dense vegetation and lower water levels
Flooded areas
Flooded areas = c4 < 0.1436 or (c4 ≥ 0.1436 & NDWIF2 ≥ - 0.5475 et DWV < -0.5092) 15
Classification accuracy and validation Classification accuracy (%) for the 3 types of marsh vegetation in Camargue:
2005
2006
Reedbeds
91,9
92,6
Club-rush
93
Aquatic beds
88,3
Acquisition in October instead of September + extremely small class ?
84,9 Aquatic beds in brackish marshes mixed with Club-rush + acquisition in October?
Classification accuracy (%) for flooded areas in 2006:
Flooded areas
All marshes
Open marshes
Vegetated marshes
76
86
70
Best results: first half of the year and reed height195 cm 23
Application for monitoring: flooding duration
Influence of water management on aquatic beds 24
Conclusion
► Remote sensing and statistical modelling for wetland monitoring : sustainability, precision, affordablility
► SPOT 5: multispectral and multitemporal modes optimal for wetland monitoring on large areas
► Roles reversed : field campaigns as a complementary tool for wetland monitoring with satellite remote sensing
25
Perspectives: improvements
► More descriptive variables : TC wetness, index differences ► Additional field campaigns to monitor reed harvesting ► Monitoring of water levels with the IME ► Number of panicles and green reeds : Rpart? GAM? ► Automatization of the methodology: simplicity for managers
26
Perspectives: other applications ► Rice cultivation:
27
Perspectives: other applications ► Rice cultivation:
PNRC: digitalization of rice fields
28
