2_04_03_143_CoconutResourceInventories

Report
Coconut Inventory Activities
in the Pacific ?
Wolf Forstreuter
Coconut Resource Known ?
• “Kiribati has
palms”
• Believable?
LINE GROUP
Atoll Name
CHRISTMAS ISLAND
FANNING ISLAND
WASHINGTON ISLAND
FLINT
CAROLINE
MALDEN
STARBUCK
VOSTOK
SUM
3,827,913 coconut
NoPlots
DEC
20 153,863
20 259,904
20 150,751
20
50,912
20
13,038
NIL
No Palms
NIL
No Palms
NIL
No Palms
SDC
164,719
8,856
798
1,218
3,978
SCC
133,722
9,794
196
1,079
2,747
Total
452,304
278,554
151,745
53,209
19,763
955,575
Palm Resource Mapping
with VHR Satellite Images
A. Separation of coconut palm from
other vegetation possible
B. Density stratification is possible
C. Counting is possible in scattered
and semi dense stands with 95%
accuracy (dense underestimation)
D. Field work necessary but reduced
to statistical minimum
Display VHR Image Data
Texture of pan-sharpened
VHR image data allows to
separate palm and natural
forest
Density Classes
A. Scattered 25 – 50 palms / hectare
B. Semi dense 50 – 150 palms /
hectare
C. Dense > 150 palms / hectare
Delineation of Palm Density
50 x 50 m grid helps interpretation
Selection of Sample Plots for
Coconut Palm Counting
1. Select are only plots that
are fully covered by one
stratum
2. Randomly selection of
about 20 % plots
3. Counting palms / plot
Counting Palms in Plots
• Placing a dot on top of every visible palm
• Digital overlay with grid and counting in
MapInfo (GIS software) dots per grid cell
• Transfer to Access
Counting in MapInfo
SQL Result
MapInfo automatically counts the
number of palms within the plots using
SQL select
Analysis in Access
•
•
•
•
•
•
•
•
Sum
Min
Max
Mean
No Plots
Area plots
Area Stratum
(Pot. No. Plots)
Mapping at SPC and PIC
SPC-SOPAC Fiji
Agriculture
Kiribati
Environment
Kiribati
Complete Kiribati and
Part of Tuvalu Mapped
Field Sample Plots Needed
• Counting palms / hectare in dense
stands
• Counting amount of hybrids
• Estimation coconut production
• Recording extent of diseases
• Estimation of palm age
• Estimation of timber volume
(diametre and height)
50 – 100% under estimation of
palms / hectare in dense stands
Field Plots
• 40 x 40 m in scattered and semi
dense stands
• Selection of image sample plots
• GPS assistance for location
identification in the field
Age / Height Measurement
• Length of 11 leave scares in cm
• Height of palm
– Distance to bottom
– Angle to bottom
– Angle to top
Fertility
• Number of coconuts / bunch
• Three oldest bunches counted
• Average calculated
Diseases
Top: Stick insect
Right: Rhinoceros beetle
Timber Volume
• DBH
• Height
• Form factor
• Timber factor
Calculation is Coded
PalmValue
Measurements
per PalmInput
CoconutBeetle
NoCoconutsB3
NoCoconutsB2
NoCoconutsB1
Length11Leave
3.67
8.33
1.67
4 3
4
7 8
2 4
1
63.67
4
5 4
4.67
4 4
44.67
6
5 6
6.33
4 6
54.67
5
56.33
8
6 6
6
4 4
4
4
10
2
2
3
3
4
3
9
4
6
6
4
StickInsect
28
29
28
29 51
28
31 42
25
25 52
35 82
25
35 91
27
40 117
32 73
25
30 87
27
35 75
26
31 75
34 74
25
34 80
22
35 70
25
Hybrid
18
18
17
22
22
19
19
15
22
20
18
16
17
Fertility RB_Attack SI_Attack Hybrid
DBH
9.09
7.92
9.35
1
-1 6.6
14.83
2
-1 6.2
3 13.64
-1 7.2
4 17.46
-0 9.2
5
-1 8.4
12.46
6
-0 12
7 12.64
-0 9
8
-1 10
12.4
9
-1 7.8
1012.16
-1 7.8
1111.47
-1 8.6
12
-1 8
9.97
13
-1 7.8
10.36
Age
DistanceDown
PalmNo
29.4
30.7
25
1
34.8
1
34.51
401
1
31.51
30.21
34.81
1
31.11
33.61
1
34
1
34.6
AngleUp
Height
PlotNo
0001_001
0001_002
0001_003
0001_004
0001_005
0001_006
0001_007
0001_008
0001_009
0001_010
0001_011
0001_012
0001_013
DBH
AngleDown
Palm_ID
TRUE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE TRUE FALSE
FALSE
FALSE
FALSE FALSE FALSE
FALSE FALSETRUE
FALSE
FALSE
FALSE
FALSE
TRUE FALSETRUE
FALSE FALSE TRUE
FALSE
TRUE
FALSE TRUE TRUE
FALSE
TRUE
FALSE FALSETRUE
FALSE FALSE TRUE
FALSE
TRUE
FALSE FALSE TRUE
TRUE TRUETRUE
FALSE
TRUE
FALSE
TRUE
TRUE TRUETRUE
FALSE TRUE FALSE
TRUE
FALSE
FALSE FALSE TRUE
FALSE
TRUE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE
Volume
0.37
0.35
0.28
0.85
0.76
1.32
0.58
0.54
0.71
0.55
0.61
0.54
0.58
Harvesting Coconuts
It is uneconomic to search for a coconut !
It is uneconomic to carry more than 300m !
Distance to Road
• Create buffer zones around tracks
and roads
• Overlay over stratified area
• Recalculate productive area
• Remote sensing, GIS and GPS
technology provide new avenues for
coconut resource inventories
• Utilising these technologies
enables quantitative estimation of
available resource
Thanks

similar documents