JARQ 29, 275-284 (1995)

Site Environments and Stand Structure of the Mangrove Forests on Pohnpei Island, Micronesia

Kiyoshi FUJIMOTO '1, Ryuichi TABU CHI'2, Tokunori MO RI'3

and Tamon MUROFUSHI''

•1 Kyushu Research Center, Forestry and Forest Products Research Institute (FFPRI) (Kurokarni, Kumarnoto, 860 Japan)

•2 Hokkaido Research Center, FFPRI (Toyohira, Sapporo, 062 Japan) •3 Bio-resource Technology Division, FFPRI (Tsukuba, Ibaraki, 305 Japan) ·• Department of Geography, Tokyo Metropolitan University (Hachioji, Tokyo, 192-03 Japan)

Abstract In order to obtain fundamental data for the prediction of stand productivity and sustainable management, two permanent plots, I ha each, were established in two types of mangrove habitats, i.e. estuary and coral reef types, on pohnpei Island, Micronesia. This paper analyzes the stand structure and the site environments and their relationships. The number of trees in the estuary type was less than half of that in the coral reef type and tree size in the former was larger than in the latter. These differences were attributed to the difference in the age between the forests which was estimated from the thickness of the mangrove peat layers. Distribution of Rhizophora apiculata andXylocarpus granatum trees may possibly be related to the sub­mergence frequency which affects the soil water content and soil water EC. Large Sonneratia alba trees were widely scattered in both plots. These trees were assumed to have survived by vegetative reproduction in spite of the changes in site environments that have occurred since the mangrove forests were developed at their present site. There were few Bruguiera gymn­orrhiza and X granatum trees with a diameter between 10 and 25 cm in the estuary type. This fact suggests that some environmental changes which disturbed the establishment of B. gymnorrhiza andX granatum trees, such as a relative rise in sea level, occurred during acer­tain period of time.

Discipline: Forestry and forest products Additional key words: coral reef type, estuary type, soil water EC, tidal condition, topogra­phy

Introduction

Pohnpei Island is located in the East Caroline Islands (lat. 6°45' to 7°04' N., long. 158°05' to 23 'E.). The island, which consists of Tertiary vol­canic rocks, is surrounded by barrier reefs except for a part of the southeastern side, where a fringing reef was formed. Mangrove forests surround the island, most of which developed on the reef flats fringing the island and some of them are situated in estuaries (Fig. I). The former is referred to as coral reef type and the latter as estuary type. There are few data on the stand structure and site environ­ments of these types of mangrove forests in com-

Present address:

parison with the delta type found in continental areas and large islands.

The USDA Forest Service estimated the timber volumes in the mangrove forests of Pohnpei Island5

>

and drew vegetation maps 3>. However, detailed investigations have not yet been conducted on the structure and site environments.

The purpose of this study is to set up permanent plots and to analyze the stand structure and site environments of each type. Furthermore, the rela­tionship between the stand structure and the site environments and the differences between the coral reef type and the estuary type will be discussed. Periodical census for the plots may enable to ana­lyze the stand dynamics. These results may contribute

• 1 Forest Environment Division, FFPRl (Tsukuba, lbaraki, 305 Japan) • 2 Shikoku Research Center, FFPRl (Asakura-tei, Kochi, 780 Japan)

276 JARQ 29(4) 1995

le:) I Coral reef

1¥m%:I Mangrove forest

158' 1 O' E

I I S8' 20' I!

I 7' 00' N-- - 7• 00' N

(>' 50' N --

158' 20 ' E

158' 10' E 5km ---•'====---Fig. I. Map showing the topography of Pohnpei Jsland, distribution of mangrove forests

and location of research plots

to the improvement of the predictions of stand productivity and to the development of a sustainable management system for mangrove forests through­out the Pacific region.

Methods

Two rectangular permanent plots, 1 ha (50 m x 200 m) each, were set up in the southeastern part of the island (Fig. 1), i.e. a coral reef type (PCJ) and estuary type (PEI). Additionally, a small plot (30 m x 60 m) also was set up at a slightly higher elevation in the estuary type (PE2). Each plot was subdivided into subplots (10 m x JO m each). Ground level was surveyed at all the grids and some addi­tional points using a pocket compass with a level in order to analyze the microtopographica.l condi­tions. The sea level when the topographical survey

was performed was used as a base level and the sur­vey data were later adjusted to the elevation using the tide table at Pohnpei Harbor.

Sedimentary structure in the mangrove habitats was examined by boring using a peat sampler in order to estimate the formative processes and age. Soil water samples were collected from several depths and their pH and EC (electric conductivity) values were measured as key environmental factors in the rhizosphere of mangrove trees. In order to observe the submergence frequency by tide fluctuation, two dataloggers were set up in PEI and PC!, respectively. One of them was set in the seawardmost part of the plots and the other in the landwardmost part of the plots.

All trees taller than 1.3 m were numbered and tagged for repeated measurement. Species, root height and stem diameter were recorded. In the case

K. Fujimoto er al.: Site E11viro11me11ts a11d Srand Structure of Mangrove Forests 011 Polmpei ls/a11d 277

of Rhizophora spp., stem diameter at 30 cm above root collar was measured, while DBH (diameter at breast height: 1.3 m) was measured for non­Rhizophora spp. Tree height was determined using a subsampling method. The positions of all the trees measured were mapped to provide further informa­tion for horizontal stand structure analysis.

Results

J) Topography of the permanent plots Fig. 2 shows the topography of the permanent

plots. PEI and PE2 were set up at almost right angles to the river channel. PEl has a tidal creek

A 8

1SO ' 16 m 40

50

20

100 11

0

0 PE2 so 6

I 0

0 SOm PE1

I I -60 - 20 0cm (a.m.s.l)

C

between line 13 and line 16. The elevation around the tidal creek is relatively low. The ground level of PE2 except for the immediate riverside area is higher than that of PEI. PCl was set up at right angles to the coastline. The ground level rises gradu­ally from seaward to landward and is higher than that of PEI. The ground level around a large Sonneratia alba tree is usually higher than that of the peripheral area, especially in PEI.

2) Sedimentary structure Fig. 3 shows the sedimentary structures in and

around the permanent plots. Mangrove peat layer overlies a marine clay layer in PEI, but overlies a

mF E D C B A 200 21

D 6

so 1SO

s

4

3

100 11

2

0

30m

50 6

I I 20 40cm (a.m.s.l)

0 PC1 SOm

- milk· I I

20 30cm (a.m.s.l)

Fig. 2. Topography of PEI, PE2 and PC! a.m.s.l.: above mean sea level.

278

Landward

Landward A 21

T I"

T I

F21

y

'r I

;r, I

A 11

IIIlIIJJ c I a y

r

PE1

0

PC 1

0

JARQ 29(4) 1995

Seaward

100m

Seaward

A1

50m

·1

-2

·3

-4

·5

-=======d I ,· ·, .. j sand !:?i:~ mangrove peat

~ humus

m peaty clay

~ shell fragment j '3 17! coral fragment

Fig. 3. Sedimentary structure around PE I and in PC! STH : Scale for tree height.

marine sand layer in PC!. As the depth of mangrove peat ranged from 3.0 to 4.8 m in PEI and 0. 7 lo 1.2 m in PC!, it is suggested that the formative age of the mangrove forest around PCJ was younger than that around PEL The formative age was esti­mated to be younger than 1,000 yrs BP for PC! and about 2,000 yrs BP for PEI on the basis of the radiocarbon data obtained from mangrove peat layers in previous studies2

•4> and the developmental processes

of mangrove habitats on Pohnpei Island 1 •2>.

3) Soil water pH and EC Soil water pH in PEI, at 6.5 - 7. I, was lower than

that in PCI which was 7.1 to 7.4 (Fig. 4). However, no difference in soil water pH within a plot was recognized. On the other hand, soil water EC decreased from seaward to landward in a plot, though around the tidal creek in PEI the value was slightly higher than at seaward sites (Fig. 5).

4) Tidal conditions The mean tidal range at Pohnpei Harbor is 70 cm,

and mean spring tidal range reaches l04 cm. Tidal fluctuations were observed using datalog­

gers between January 24 and February 28 in I 994 at the seaward site of PEI, between February 7 and February 28 at the landward site of PE I , between January 20 and February 28 at the seaward site of

7.5 r----_, ---7 -- .---_./.' ,. ...... __ ..,,,,.. .... -""'~.:--- _/-s<~ )--; -I)- ..: := =: :: ~~ /

..,,./ -..:-, ';., ..... /

6.5

6 -- PEI Depth PCl Dcpch --o- A:20cm ..... A:20cm . A:40cm . -0- -- A:40cm -- -. -lr- F:20cm ..._ F:20cm

s.s .

--9--- F:40cm ...._ F:40cm

' . s 6 16

Seaward 11

Linc No.

21 Landward

Fig. 4. Soil water pH in PEI and PC!

K. Fujimoto et al.: Site f:11viro11me11ts and S1a11d Slruc/ure of Mangrove Fores1s 011 Poh11pei Js/011d 279

45

40

'? 35 .!l (I)

8 w 30 PEI Depth -o- A:20cm

25 --- A:40cm -6- F:20cm -- F:40cm

20 1 6 11 16 21

Seaward Line No. Landward

so 45

40

'? 35 ~ (I)

30 8 w 25

20

15

10 1 6 11 16 21

Seaward Line No. Landw:Jlrd

Fig. 5. Soil water EC in PEI and PCI

PCI, and between January 21 and February 7 at the landward site of PCI. Every high tide level at the landward sile was lower than that at the seaward site in both plots. The average decrease in the tide level during the observation of tide was 16 cm in PEI and 12 cm in PCl. The decrease in the values was div.ided between all the points surveyed of the ground level in proportion to the distance from line 1 where the seaward dataloggers were set, and the minimum high tide level for submergence at every point was estimated using the decreased value. The submergence frequency at every point was obtained by counting the frequency of high tide events over the minimum level using the tide table of 1994 and was expressed in Figs. 6, 7 and 8 as the ratio 10

the total high tide frequency in 1994. In most of the area in PEJ, the submergence fre­

quency exceeded 900Jo except for the landwardmost part and the area around some large Sonneratia alba trees (Fig. 6). On the other hand, for most of the area in PE2 the frequency was below 50% except

for the seawardmost part (Fig. 7). In PCI , it decreased gradually from over 70% in the seaward area to below 40% in the landward area (Fig. 8).

5) Stand structure Total numbers of trees measured at PEI, PE2

and PC! were 651, 151 and 1,554, respectively. Fig. 9 shows the frequency distribution of the diameter of the respective species in PEI and PCl .

In PEI , Rhizophora apiculata, Bruguiera gym­norrhiza, Xylocarpus granatum and Sonneratia alba were observed. Numbers of these species were 369, 184, 64 and 34, respectively. Few of the R. apiculata had a diameter below 10 cm. On the other hand, many B. gymnorrhiza with a DBH below JO cm were observed though in few trees the DBH ranged be­tween 10 and 25 cm. There were relatively few X. granatum trees with a DBH below JO cm and be­tween 15 and 25 cm. The DBH of S. alba was gener­ally large though the number of trees was small, and none showed a DBH below 20 cm. The DBH of the largest tree in this plot, S. alba, was 204 cm. R. apiculata was distributed throughout the plot. However, trees with a diameter between 10 and 30 cm were concentrated in the area where the submer­gence frequency exceeded 90% (Fig. 6). B. gymnor­rhiza and S. alba were equally distributed in the plot. X. granatum was not found around the tidal creek where the submergence frequency was high.

In PE2, R. apiculata, 8. gymnorrhiza, X. grana­tum, Lumnitzera littorea and Heritiera /iuoralis were observed (Fig. 7). The respective numbers of trees were 61 , 23, 57, 7 and 3. S. alba was not found in the plot. The frequency distributions of R. apiculata and 8. gymnorrhiza were almost the same as those of PEI. However, there were many small X. granatum trees with a DBH below 10 cm and some large trees with a DBH between 75 and J35 cm.

In PCl, R. apiculata, B. gymnorrhiza and S. alba were observed, and the numbers of these trees were I, 168, 352 and 34, respectively. X. granatum was not detected. This plot was dominated by dense stands of R. apiculata, particularly in a landward quarter of the plot and the seaward part where the submergence frequency was high (Fig. 8). Few trees showed a diameter below 10 cm. On the other hand, many small B. gymnorrhiza with a DBH below 10 cm were distributed in this plot except for a landward quarter of the plot. Some large S. alba trees were scattered throughout the plot, though there were no small S. alba trees with a DBH below 20 cm. Tree size was smaller than that in PEI (Fig. 9).

mA 8 C - -200 d I I ..... ~

150

100

.. • • • • • • • •

,.. .. _., .. .,.~ . . . ., .. •• •• ... "• . ;;. . :-:. • 9 } •.••. 4 . ·· --. . .. . . ~·. :.·~· .. '• . ~·· ., ,. .. •.. .. ... . .. ·. ·. . . • ... ~ ~· .. . · ... ··J·•.... . ... . . : .: .... .

• •.• • ... • u.• ••••• ,,._:, ~·-·· ( .. .. .. . . II

ii• . • . -. " .. . . •• i .. . : . . ..• ; i • • ~ J • . ~ t • • . • •

·: .. ~ ~. . .. • · ... . .~ ...... .

11

~ so C 6

o : : : : r , F'1~ 1 0

Rhizophora 50m

mA 8 C D E F mA 8 C D E F mA 8 C D E F 200 41 I I I ,,..,. I i 21 200 21 200 ii I / \ I I r I 21

150

100

• •

,; • •• •.;.1 ~ • •

• •

•

•· • • • •

•

•

~

16 150

11 100

16 150

or.. .: 11 100

I . •

. . • • • I . ,,/ . ., .

.. ; • C • e ... •' Wllllft ._ .,,. ___ ,._.

•• ., D

~ i

I • • ::;:·:

• •

~

16

11

50~6

. -~ • . •iit · . 50~6 50~6

I .... II -. ·. ;, .,· . •:' . ; ... ,. .IJ · . • ;p. f

'. . .·~ •..

·<, •

~;· ··\ 0 1

0 50m Bruguiera

PE1

0

80

- • . ,.,, I . ' . :itlit • :

I~ if· ,• I I ;If -., .. •"' !"'' • .• ~

1 0 Som

Xylocarpus

60%

o I 0

Fig. 6. Submergence frequency and distribution of each species in PEI

I

,•it

111: ·Rt .:a, -"'-~ -

J\ J.rf I I 1 50m

Sonneratia

"' ~

Diameter(cm)

o,so-0100-1 so 0 so-100 0 40-50 0 30-40 0 20-30 0 10-20 0 0-10

... ;i:,. ;:c, t:)

"' \0

:5 ~ \0 V.

'~L i ~1 r ~1~: J .1 ·• r ,~· · C

06

• . 1. e• 1 •

I I I I I Diameter(cm)

• I I I I I - I I - I I • 100-

• 50-100 • 40-50 e 30-40

0 I I J 1 0 I I .r 1 0 ,. - 19 .. sr 1 O+---t-rlilllllll • 20-30 • 10-20

O 30m o 30m O 30m O 30m

Rhizophora Bruguiera Xylocarpus • Lumnitzera o Heritiera

PE2 !IIIIIIIIR[wrL~~ Ii I I

60 40%

Fig. 7. Submergence frequency and distribution of each species in PE2

• 0-10

~

:!l ~-!" 0 ~ ,, :-

~ r:, ~ a· :,

~ ;;. 0,

t ~ ~

r ,:: ~

~

~ i ~

~ l)l .:;

g ~ :,.

'i .. ;;, iS" ~

.., 00

282 JARQ 29(4) 1995

mF E D C 8 A 200 --;;;,-..,..--,-...,..---,21

Rhizophora

PC1

50m Bruguiera

I 60

•

6

SOm

I

40%

Sonneratia

Olameter(cm)

e 100-1so • so-100 e 40-SO • 30-40 • 20:...30 • 10-20 • 0-10

fig. 8. Submergence frequency and distribution of each species in PCI

Discussion

I) Relationship between stand structure and site environments

(I) Differences between the estuary type and the coral reef type

The number of trees in PEI was less than half of that in PCI, and the tree size in PE ! was larger than that in PC!. These differences were attributed to the difference in age between the forests. The formative ages of the forests around PC! and PEI were estimated to be younger than 1,000 yrs BP and about 2,000 yrs BP, respectively. The fact that Xylocarpus granatum was not detected in PC! may also be caused by the difference in forest age.

( 2) Distribution of each species and site environ­ments in a plot

Relative frequency of appearance of Rhizophora apiculata was high in the area where the submergence frequency was also high in every plot except for the landward part of PCI. On the other hand, Xylocar­pus granatum occurred where the submergence fre­quency was relatively low. Usually, the submergence frequency affects the water content and chemical con­ditions of the soil. A proportional relationship was recognized between the submergence frequency and soil water EC in PEI and PC! (Figs. 5, 6 and 8), though the soil water pH was not related to the sub­mergence frequency (Fig. 4). Distributions of R. apicu/ata and X. granatum may possibly be related to the soil water content and soil water EC. In the

K. F11Ji111oto et al. : Site Environments and Sttmd Stmct11re of Mangrove Forests 011 Pohnpei Island 283

120 100

>, g 80

I Go it 40

20

-

!II Q O 20 40

so 40

~ g 30 ::, l 20

10 --00 •• 20 40 60

Ml. II ~111 uzop 1ora-

030C (cm)

LI 111 J 11 Brtt mera-

OBH (cm) 10~~~~~~~~...,..,...,..,..,......,~.,......~..,..,...,..,...m....,

>, 0 C

"' ::,

[ u.

8 -14-a!-l-lll-!-l..l-l-l+l++-H-IH-++l-l+l++-H

6

4

2

0 0 20 40 60 OOH (cm)

3 -.-,...,..,...,.,.,,.,._,..,......,...,.....,...,.,...,...,.....,..,..rr,"TT"T"T""r"TT"TT.,..,...l"T1""T"I

2

0 0 PEI

50 100 150 DBH (,cm)

--

500

400

300

200

100 : .• 00 200

150

100

so

20 40

--

lWli 20 40

k~l.11~11 1 IIZOP, 1qra

030C (cm)

k~l.1,~Je!J ~

.. - 1-1-

DBl-i (cm)

5 ...,.,...,..,...,.,.,_,.,,.,..,..,..,.,..,...,..,...,...m..,..,..,......,...,..,...,..,..m-,-,...-.-mc-,,

4 +t++H-1-.Hill+t-H-t+H-t-H++H-l+++hl,

3

2

1

00 PC! so 100 150 06H (cm)

Fig. 9. Frequency disLribution of diameter of respective species DBH: Diameter at breast height, D30C: Diameter at 30 cm above root collar.

landward part of PCI, R. apicu/ata with a diameter between 10 and 30 cm dominated despite the low submergence frequency and the low soil water EC. Usually, under such circumstances, if the forest is mature and has no Xylocarpus spp., the relative frequency of Bruguiera gymnorrhiza ought to be high. It is difficult to explain, based on pure.ly natural factors, the reason why R. apicu/ata was a dominant species in such site environments. There­fore, human activities may have affected the forma­tion of the stand structure.

2) Factors related to the size structure of respective species

It is worth noting that only large trees of Sonner­atia alba were widely distributed in both PEI and PCl. This phenomenon has been recognized in other mangrove forests on this Island 1>. Usually, S. alba can be established only at an open site such as in the front of a forest as a pioneer species. lt is

assumed that these trees were able to survive by vegetative reproduction in spite of the changes in site environments since the mangrove forest deve­loped at its present site.

There were few Rhizophora apiculata trees with a diameter below IO cm in all the plots; on the other hand, there was a large number of small Bruguiera gymnorrhiza trees in this diameter class. This trend has been generally recognized in mangrove forests. However, it is worth noting that lhere were few B. gymnorrhiza trees with a diameter between 10 and 25 cm in PEI. Size structure of Xylocarpus grana­tum in PEI showed the same tendency, suggesting that some environmemal changes which disturbed the establishment of B. gymnorrhiza and X. granatum had occurred during a certain period. The establish­ment of B. gymnorrhiza in PC! after the occurrence of environmental changes may account for the fact that this tendency was not recognized in PC! and the relative rise in the sea level could be one of

284

Geomorphological situation Sedimen1a1ion in estuary Fonnn1ion of coral reef

JARQ 29(4) 1995

141--./. Sea-level change

Vege1a1ive reproduction of So1111eratia alba

,.---- ------, I Differences in tree density, size 1

of tree and species structure I 1 between different habitat types

I - ----- --,

I Distr ibution and size structure 1--. ~ • of each species in a plot _ J~

L----------'

M icroiopography Decremer11 of 1idal level 1----~~ in mangrove fores!

Fig. JO. Site environments comrolling formation of stand structures of mangrove forests on Pohnpei Island

the causes. Fig. 10 shows the site environments controlling

the formation of stand structures of the mangrove forests on Pohnpei Island as a conclusion of this paper.

References

I) Fuj imoto, K. & Miyagi, T. (1993): Development process of tidal-flat type mangrove habitats and their zonation in the Pacific Ocean: A geomorphological study. Vegetation, 106, 137-146.

2) Fujimoto, K., Miyagi, T. & Kikuchi, T. (1995) : For­mative and maimainable mechanisms of mangrove habitats in Micronesia and the Philippines. /11 Report of grant-in-aid for scientific research from the Minis­try of Education, Science and Culture, Japan, No.04041020: Rapid sea level rise and mangrove

habitat. ed. Kikuchi, T., 9-18. 3) MacLean, C. D. et al. (1986): Vegetation survey of

Pohnpei, Federated States of Micronesia. Resour. Bull., PSW-18, Pacific Southwest Forest and Range Experiment Station, Forest Service, U.S. Department of Agriculture, Berkeley, CA, 8pp. + 11 maps.

4) Miyagi , T. & Fujimoto, K. (1989): Geomorphological situation and stability of mangrove habitat of Truk Atoll and Ponapc Island in the Federated States of Micronesia. Sci. Rep. Tohoku Univ., Ser.7 (Gcogr.), 39, 25 -52.

5) Petteys, E. Q. P. et al. (1986): Timber volumes in the mangrove forests of Pohnpei, Federated States of Micronesia. Resour. Bull., PSW-19, Pacific South­west Forest and Range Experiment Station, Forest Service, U.S. Department of Agriculture, Berke.Icy, CA, 2pp.

(Received for publication, Dec. 27, 1994)