numerical study of intertidal zonation along visakhapatnam...
TRANSCRIPT
-
Indian Journal of Marine SciencesVol. 15. June 1986. pp. 102-106
Numerical Study of Intertidal Zonation along Visakhapatnam Coast, EastCoast of India
G M NARASIMHA RAO & M UMAMAHESWARA RAO
Department of Botany. Andhra University. Waltair 530003
Receired i') February 1985; revised •.eceired 27 January 1986
Quantitative data were collected for I y (Jan. to Dec. 1983) to study vertical distribution of plants and animals on intertidalrocky surfaces. CoelTicients of community 'Similarity and cluster analysis were used for studying the distribution of intertidalpopulations. Present data clearly showed the occurrence of 3 zones in the intertidal region of Visakhaparnarn.
Intertidal zonation was studied in India fromVisakhapatnam 1.2, Mandapam ", Okha":" andAnjidiv Island". The zonation patterns and theclassifications followed for describing zonation variedin the above areas. Russell:" applied numericalmethods to study the distribution of populations andstatus of different zones present on intertidal rockysurfaces. Using these analytical methods, the presentstudy was undertaken to examine the zonation ofVisakhapatnam coast.
Materials and MethodsVisakhapatnam (I" 14'30" and 11"45' N; 83" 16'25"
and 8321 '30" E) coast is sandy with outcrops of rockyboulders of various sizes and shapes. Quadrat sampleswere collected from 4 stations. selected previously",and from another station at Jodugullapalem. situated4 km north of Visakhapatnam.
Algae and animals occurring at different verticallevels on the rocky surfaces were sampled7.s at randomusing a quadrat (aluminium frame of 0.5 x 0.5 m insize). The number of algae and animals present in eachquadrat was noted and cover values were estimated foras many species as possible. For estimating the cover,the quadrat frame was sub-divided with nylon threadinto 25 sub quadrats and the cover data of the wholequadrat were expressed as percentage on a 5 point scale(I = 0-20; 2 = 21-40; 3 = 41-60; 4 = 61-80; 5 = ~I-100). A second quadrat sample was taken by placingthe aluminium frame either to the right or left of thefirst quadrat as suggested by Russellx. During thestudy period (Jan. to Dee. 1983) 144 quadrat sampleswere collected.
Species area curve was plotted using the formula 9
(1 - ~ ) q where N = total number of quadrats, n =102
number of species present in the total number ofquadrats. q == serial number of quadrats. The numberof predicted species present in the quadrats increasedrapidly from 2nd quadrat onwards up to 60th quadratand there after remained constant indicating that asample size 60 quadrats is necessary for obtainingmaximum number of species in the area investigated.
Population similarity coefficients of Jaccard 10.Sorensen 1 1 and G leasen 12 were used to compare thequadrat samples collected from EHWS (2 m CD) toELWS( - 0.4 m CD) in the intertidal region. The first 2coefficients are based on species presence data and thethird one on the cover data. Mean values of duplicatesamples were used for calculating the similaritycoefficients. The I value (0.8055) calculated for the 2sets of quadrat samples was less than the table value at5~
-
RAO & RAO: NUMERICAL STUDY OF INTERTIDAL ZONATION
ResultsRelative frequency of 50 species obtained in the
quadrats varied from 4.8~;;;(Ceramium cruciatumi to36.1 % (V/va fasciatat. Data obtained on quadratsimilarity are presented in Figs 1 to 3. In the similaritymatrix based on coefficients of Sorensen 11 , 3 differentblocks can be seen (Fig. I). Littorina undulata,Nodilittorina millegrana are the species of the uppershore block. Algae were not found in the quadratsamples of this upper shore block. The speciesobserved in many quadrats of the middle shore blockwere: algae-Enteromorp'ha compressa, Chaetomorphaantennina, Porphyra »ietnamensis, Po/ysiphoniaferulacea, Acrochaetium iyengarii, Liagora erecta,
Peyssonnelia conchicola; and animals-Chthamalusstellatus, Crassostrea gryphoides, Cellana radiataradiata, Nerita a/bicilla, Morula granulata, Balanustintinabulum tintinabulum . The species observed inquadrats of the lower shore block were: algae-Cladophora utricu/osa, C. socialis, Spongomorphaindica, Bood/ea struoeoides, Bryopsis pennata, Caulerpafastigiata, c. racemosa, C. sertularioides, C. taxifolia,Dictyota aichotoma, Padina tetrastromatica, Sar-gassum vulgare, Amphiroa fragilissima, Jania rubens,Grateloupia lithophila, Gelidiopsis oariabilis, Graci/ariacorticata, G. textorii, Hypnea oalentiae, Gigartinaacicularis, Wrangelia argus, Aglaothamnion corda turn,Ceramium cruciatum, C. fimbriatum, Bryocladia
Key to Similarity percentages
o 0-20~ 21-1..0§ 1.1-6000 61-80• 81-100
10
20
30
40
50
x xx
xX
x
X
60>0< X>0< X>0
-
INDIAN J MAR SCI. VOL. 15. JUNE 1986
thwaitesii, Lyngbya majuscula; and animals-Electrapi/osa, Stegmoporel/a buskii, Stomopneustes sp. andGemmaria sp. The quadrat samples of upper shoreblock showed complete dissimilarity with theremaining quadrats of middle and lower shore blocks.Similarly the quadrats of middle shore block and lowershore block showed dissimilarity with remainingblocks, because of the differences in the speciescomposition. These observations on the similaritybetween quadrats collected at different vertical levelsindicate that 3 zones with sharp boundaries betweenthem occur in the intertidal region of theVisakhapatnam coast. The similarity matrix based on
Jaccard 10 coefficient is shown in Fig. 2. As observed inSorensen 11 matrix, strong internal similarity was seenwithin the 3 blocks. The segregation of quadratsamples into 3 blocks or groups is very clear inJaccard 1 0 coefficient. Matrix based on Gleason 12coefficient is shown in Fig. 3. Clear-cut demarcation ofthe 3 zones can be seen in this matrix prepared fromspecies cover data. But the similarity values withineach block varied, depending on the cover data of eachspecies and weightage given to it.
Overlapping was observed between the 3 differentblocks (Figs I and 3). This may be due to sampling onirregular rocky boulders: downward and upward
104
Fig. 2 -Similarity matrix of quadrat samples using Jaccard coefficient (Similarity percentages as in Fig. I)
-
RAO & RAO: NUMERICAL STUDY OF INTERTIDAL ZONATION
movements of animals and also due to settlement oforganisms depending on the differences in micro-habitats. The similarity values obtained for thequadrats of overlapping populations were > 20%when Sorensen 11 and Gleason 12 coefficients were used(Figs 1 and 3). On the other hand with Jaccord+"coefficient the similarity values were < 20% and hencethey were not depicted in the matrix (Fig. 2).
From cluster analysis(Fig. 4) 3 different groups haveemerged from the populations of the intertidal region.Of the 50 species, recorded in 144quadrats, 18 specieshad more than the predicted value at 0.001 probability(Fig. 4) and the 3 distinct clusters obtained in the
10
20
30
40
50
present study clearly show the occurrence of 3 zones inthe intertidal region of Visakhapatnam.
DiscussionAt Visakhapatnarrr' , the intertidal region is divided
into 3 zones following the classification of Stephensonand Stephenson 13. The upper most supralittoralfringe extends from about 1.4 m CD to extreme highwater springs (2 m CD). Algae were not found in thiszone and it can be recognised by the presence ofLittorina sp . The midlittoral zone is well defined withan average width of 1 m (0.4 to 1.4 m CD). There is adistinct barnacle line formed by Chthamalus stellatus
x X
60XX XX
>0< X
7072
10 20 30 40 50 60 7072Fig. 3 -Similarity matrix of quadrat samples using Gleason coefficient (Similarity percentages as in Fig I)
105
-
INDIAN J MAR SCI, VOL. 15, JUNE 1986
ISO140
100
80
bO
40
bFig. 4 +Cluster analysis based on highest 'r values of speciespresence in 144 quadrats [Species: (a) I Littorina undulata, 2Nodilitt orina millegrana: (b) 3 Chthamalus stellatus, 4 Crassostreagraphoides, 7 Morula granulata, 10 Ulra fasciata, 11 Enteromorphacompressa, 12 Chaetomorpha antennina: and (e) 26 Spongomorphaindica, 27 Boodlea struveoides, 37 Wrangelia argus, 38 Graci/ariacorticate, 41 Hypnea ralentiae, 43 Amphiroa [rugilissima, 44 Janiarubens, 45 Chaetomorpha brachygona, 48 Stegmoporella husk ii. 50
Gemmaria spp]
at 1.4 m level at the upper limit and a clearcut break inthe vegetation at the lower limit of midlittoral zone.The infralittoral fringe extends from 0.4 down to - 0.2or - 0.4 m CD (El, WS). From the present analysis onthe similarity between quadrat samples (Figs 1 to 3)and on the association between species (Fig. 4), it isclear that 3 different communities occur in theintertidal region of the Visakhapatnam coast fromhigh water to low water level, with strong internalsimilarity between species of each community anddissimilarity with other communities of species. Theupper block in Figs 1 to 3 (quadrats 1-20) and cluster'a' in Fig. 4 correspond to the populations of supralittoral fringe, the middle block (quadrats 21-46) andcluster 'b' to populations of midlittoral zone (from 0.4to 1.4 m CD) and lower block (quadrats 47-72) andcluster 'c' to populations of infralittoral fringe (0.4 to- 0.2 or - 0.4 m CD).
At Visakhapatnam, Rao and Raol sub-divided themidlittoral zone into 3 sub-zones, viz, upperlittoral,mid littoral and lowerlittoral. The dominant and bandforming algae like Sargassum spp and Graci/ariacorticate were observed by them in the lowerlittoral aswell as in the infralittoral fringe and they suggested
106
that zonation pattern conformed with the classi-fication proposed by Womersley and Edmonds!".UrnamaheswaraRao and Sreeramulu ' didn't observeany subdivision and Viva fasciata, Chaetomorphaantennina, Enteromorpha compressa, Porphyra viet-namensis, Bangiopsis subsimplex, Chthamalus stellatuswere found in the entire midJittoral zone from 0.4 to1.4 CD as continuous bands. In the present study alsototal internal similarity (Figs 1 to 3) and positivecorrelation (Fig. 4) were observed between thequadrats sampled and populations of the midlittoralzone and dissimilarity and negative correlation werefound between the quadrats and populations ofmid littoral zone and infralittoral fringe (Figs 1 to 3).These findings on population similarity and clusteranalysis further, confirm the observations ofUmamaheswara Rao and Sreeramulu".
Russe1l7.8 used coefficients of Sorensen! I andGleason 12 in 'his quantitative work on zonation. In thepresent study in addition to Sorensen II and Gleason 12coefficients, Jaccard 1 0 coefficient was also used forspecies presence data and it eliminated the smallpercentage of overlapping populations observed insome quadrats of the 3 zones. When compared withSorensen and Gleason coefficients, Jaccard coefficientappears to be more suitable for the analysis ofintertidal communities.
AcknowledgementThe authors are grateful to Prof G Russell,
University of liverpool, UK, for going through themanuscript and offering valuable suggestions.
References1 Rao T S S & Suryananda Rao V V, J Zoo/ Soc India, 14 (1962)
112.2 Umamaheswara Rao M & Sreeramulu T, J Ecol, 52 (1964) 595.3 Umamaheswara Rao M, Proc Indian Nat Sci Acad, 388 (1972)
298.4 Gopalakrishnan P, J Mar Bioi Assoc India, 12 (1970) 15.5 Murthy M S, Bhattacharya & Radia p, Bot Mar, 21 (1978) 382.6 Parulekar A H, Proc Indian Nat Sci Acad, 39 (1973) 611.7 Russell G, J Ecol, 60 (1972) 539.8 Russell G, Oikos, 24 (1973) 158.9 Russell G, (personal communication).
iO Jaccard p, Bull Soc Weed Sci Nat, 38 (1902) 69.11 Sorensen T, Bioi Skr, 5 (1948) I.12 Gleason H A, Bull Torrey Bot Club, 47 (1920) 21.13 Stephenson T A & Stephenson A, J £Co;' 37 (1949) 289.14 Womersley H B S & Edmonds S J, J Eco/, 40 (1952) 84.