Spatial dynamics of teak defoliator (Hyblaea puera Cramer) - Cochin ...

SPATIAL DYNAMICS OF TEAK DEFOLIATOR 

(HYBLAEA PUERA CRAMER) 

OUTBREAKS: PATTERNS AND CAUSES 

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF 

THE REQUIREMENTS FOR THE DEGREE OF 

DOCTOR OF PHILOSOPHY 

OF THE 

COCHIN UNIVERSITY OF SCIENCE AND TECHNOLOGY 

By 

T.V. SAJEEV, M.Se. 

DIVISION OF ENTOMOLOGY 

KERALA FOREST RESEARCH INSTITUTE 

PEECHI, 680 653, KERALA 

SEPTEMBER 1999 

G78b33

DECLARATION 

I hereby declare that this thesis entitled "Spatial dynamics of teak defoliator 

(Hyblaea puera Cramer) outbreaks: patterns and causes" has not previously formed the 

basis ofany degree, diploma, associateship, fellowship or other similar titles or recognition. 

Peechi 

27 th August 1999. 

T.V.Sajeev

CERTIFICATE 

This is to certify that the Ph.D thesis entitled "Spatial dynamics of teak 

defoliator (Hyb/aea puera Cramer) outbreaks: patterns and causes" is a genuine record of 

the research work done by Shri. T.V.Sajeev (Reg.No. 1459) under my scientific supervision 

and the work has not formed the basis for the award of any degree, diploma or 

associateship in any University. 

Thiruvananthapuram 

28 th August 1999 

(Dr. K.S.S.Nair) 

Supervising Guide

ACKNOWLEDGMENTS 

I wish to place on record my deep sense ofgratitude to: 

Supervising guide, Dr. K.S.S.Nair, former Director, KFRI for suggesting 

this problem, his guidance and creative suggestions. 

Dr. R.V.Varma, Scientist-in-Charge, Entomology Division, KFRI for his 

encouragement and kind reminders on the task at hand. 

My colleagues, Dr.V.V.Sudheendrakumar and Dr. K.Mohandas, for their 

work on the teak defoliator which laid down the framework within which this 

work was possible. 

Dr.P.S.Roy, Head, Forestry and Ecology Division, Indian Institute of 

Remote Sensing, Dehra Dun, for extending the facilities for the initial GIS 

anaysis. 

Dr.A.R.R.Menon, Scientist, Ecology Division, KFRI, for providing aerial 

photographs and topographic sheets for the study. 

Dr. Francois Houlier, former Director, French Institute ofPondicherry, Dr. 

Sonia Darraccq, and Filiduaro Limaire, Geomatics Division, French Institute of 

Pondicherry for extending facilities for the final GIS analysis. 

Mr. Santhosh K. John, former Officer-in-Charge, KFRI Subcentre, 

Nilambur, for providing infrastructural facilities at Nilambur and for his keen 

interest in the topic.

Mr. Rajan James, Mr. A. Moosa, Mr. K. Sunny, Mr. Saji John and Mr. K. 

Mohammed ofKFRI Subcentre, Nilambur for their painstaking efforts in assisting 

data collection. 

Mr. Byju Hameed and Mr. K.V.Sathyakumar, Research fellows, KFRI, for 

their helpful participation in field trips. 

at Nilambur. 

Mr. K.P.Manoj, KFRI Subcentre, for his cheerful hospitality while being 

To my parents, and all my dears and nears who have extended moral 

supportto see me through the completion ofwork.

Chapter 

I. Introduction 

11. Review ofliterature 

Ill. General methods 

3.1. Introduction 

3.2. Study area 

TABLE OF CONTENTS 

3.3. Preparation ofplantation maps 

3.4. Observations at Kariem Muriem 

3.5. Observations in the entire Nilambur teak plantations 

IV. Studies on the spatial distribution ofoutbreaks 

in Kariem Muriem teak plantations at Nilambur 

4.1. Introduction 

4.2. Methods 

4.3. Results 

4.2.1. Preparation ofmaps 

4.2.1.1. Outbreak maps 

4.2.1.2. Elevation and aspect maps 

4.2.2. Spatial autocorrelation analysis 

4.2.3. Correlation between outbreak incidence 

and topographic features 

4.3.1. Outbreak pattern 1992 



4.3.4. Frequency ofoutbreaks in space 

Page 

1 

5 

11 

11 

11 

12 

12 

14 

15 

15 

16 

17 

17 

18 

18 

19 

19 

19 

22 

25 

25

4.3.5. Correlation between outbreak 

4.4. Discussion 

incidence and topographic features 

V. Studies on the spatial distribution ofoutbreaks 

in the entire teak plantations at Nilambur 38 

5.1. Introduction 38 

5.2. Methods 38 

5.3. Results 39 

5.4. Discussion 42 

VI. Field observations on moth behaviour 69 




6.3.1. Post-emergence behaviour 71 

6.3.2. Aggregation 72 

6.3.3. Flight behaviour and dispersal 73 

6.3.3.1. Movement ofmoths outside the plantation 73 

6.3.3.2. Movement ofmoths within the plantation 74 

6.3.4. Oviposition 75 


VII. Development ofoutbreak monitoring methods 78 





30 

35

VIII. Population dynamics ofHyblaea puera 

a synthesis ofavailable information 86 

8.1 Introduction 86 

8.2 Aggregation and flight ofmoths 87 

8.3 Origin ofoutbreaks 88 

8.4 Spatial spread ofoutbreaks 88 

8.5.An explanatory model for the population dynamics 

ofHyblaea puera 89 

8.5.1 The background 89 

8.5.2 The model 92 

References 95 

Appendix A : Algorithm for computation ofautocorrelation indices 101 

Appendix B : Light-trap data in relation to incidence ofoutbreak and 

local moth emergence 102

CHAPTER I 

INTRODUCTION 

Insects are of interest to man because of two major reasons- the large diversity 

exhibited by the group, and man's conflict with insects for food and other 

resources. Insects that draw our attention because of the latter reason are called 

pests. In most cases, being a pest is a matter of abundance of individuals. At some 

times, the population density increases to cause economic damage but at other 

times the density remains low. This increase and decrease ofpopulation density of 

different species of insects has intrigued the population biologists for long. 

Although it is generally agreed that this fluctuation is related to the availability of 

food and other resources, other factors such as natural enemies and climate are 

also thought to regulate the size ofa population. 

The teak defoliator, Hyblaea puera Cramer (Lepidoptera, Hyblaeidae) 

which is recognized as a serious pest of the teak tree (Tectona grandis L.f..) is a 

typical pest that exhibits this characteristic shift in population density. Teak is a 

multipurpose timber species, which naturally occurs in India, Myanmar, Thailand 

and Laos. During the last century when the natural teak stands could not cater to 

the needs, plantations of teak became a necessity. The first plantations in India 

were established at Nilambur (Kerala) during 1842-44. Since then, the area under 

teak has steadily increased and plantations have been raised in several other 

tropical countries of the world as well. Presently, in Kerala, the teak plantations 

extend to an area of about 78,800 ha (Shanmuganathan, 1997). This is 46.5% of 

the total area under forest plantations in Kerala. The major teak growing areas in 

Kerala are Wayanad, Nilambur, Parambikulam, Nelliampathy, Achenkoil, 

Aryankavu, Konni, Ranni and Malayattoor. 

Although 171 species of insects are recorded as associated with teak, only 

a few have attained pest status (Beeson, 1941). These are the white grubs, which 

attack seedling in the nurseries, the sapling borer (Sahyadrassus malabaricus)

(Nair, 1987), the trunk borer (Alcterogystia cadambae) (Mathew, 1991), the teak 

skeletonizer (Eutectona macheralis) (Beeson, 1941), and the teak defoliator 

(Hyblaea puera) (Beeson, 1941). While the white grubs are usually found 

restricted to the nurseries, the sapling borer to young trees; the trunk borer to 

specific regions; and the skeletonizer, to a period in the year when teak is about to 

shed its leaves, the teak defoliator occurs in almost all teak plantations during the 

active growing period of the tree. This characteristic makes it the most serious 

pest of teak. It has been estimated that damage caused by this insect in 4-8 year 

old plantations leads to an increment loss of3 m 3/ha/year (Nair et al, 1996). 

Realizing the economic loss caused by the insect, attempts were made in 

the past to standardize control methods, which included biological control as well 

as aerial spraying of chemical insecticides. Biological control using insect 

parasitoids did not prove successful and the environmental impact of insecticides 

makes them unsuitable. Biological control using a recently identified Nuclear 

Polyhedrosis Virus (NPV) (Sudheendrakumar et al, 1988) is seen as a promising 

alternative because it is quick acting (Nair et al, 1996) and is highly specific to 

this insect. 

However, the major problem in using any control agent is the difficulty in 

detecting the teak defoliator outbreaks early enough to apply the control measures. 

The vast extent of the plantations and the hilly terrain in which the search has to 

be made to detect the early sites of outbreaks pose practical problems. The larval 

life span of the insect lasts only for about 15 days within which the entire foliage 

on the tree may be eaten off. To prevent damage, the early instars of the larvae 

have to be detected and controlled. The sudden appearance of infestations in 

widely separated patches during the early outbreak period, suggests that 

successful control of the pest could only be achieved by understanding the 

population dynamics ofthe insect. 

Recent research has shown that the population trend of the teak defoliator 

exhibits several distinct phases (Mohanadas, 1995). The first phase which start 

2

covering the entire plantations of Nilambur in one year. Behavioural studies on 

defoliator moths are given in Chapter 6. Attempts to develop monitoring 

techniques to detect outbreaks are presented in Chapter 7. In Chapter 8, an attempt 

is made to synthesize all available empirical information on the population 

dynamics of Hyblaea puera in the light of recent advances in theory. The 

literature referred to for this work is presented next. as per guidelines given by 

Anderson et at (1970). The algorithm used for the computation of auto correlation 

indices is given in Appendix A and light-trap data in relation to incidence of 

defoliator outbreak and local moth emergence is given in Appendix B. 

4

CHAPTER11 

REVIEW OF LITERATURE 

The teak defoliator was recognized as a pest of teak in India as early as 1898 

(Bourdillon, 1898). During the past 100 years, the major topics of interest were 

the impact, biology, ecology, natural enemies, and control strategies related to this 

insect. 

A general description of the different life stages of teak defoliator was 

presented during the early part of the century (Stebbing, 1903). Two different 

species were described, Hyblaea puera Cramer and Hyblaea constellata Guen. 

along with a variety described as The Black Hyblaea- Hyblaea puera var. nigra. 

Descriptions of all the above said insects resembled each other except for the 

colouration in larvae and adults. Distribution of Hpuera and Hpuera var. nigra 

was continuous throughout India and Burma while Hconstellata was recorded 

only from Burma. In the next year (Hole, 1904) it was reported that Hconstellata 

differed from Hpuera with respect to two characters which are (a) H constellata 

has the outer margin of the forewing excised below the apex and excurved at the 

centre, whereas in H puera the margin is evenly curved and not excised, and (b) 

in H constellata, in the anal angle, on the under side of the hind wing, there is a 

single black spot, whereas in H puera there are two such spots. He commended 

that it is untimely to regard H.puera var. nigra as distinct from Hpuera. The 

present study does not consider the variety nigra as distinct from Il.puera. All the 

three insects were grouped under the family Noctuidae until Zemy and Beier 

classified the genus Hyblaea Fabricius under the family Hyblaeidae in 1936 

(Singh, 1955). 

Early research in Burma (Mackenzie, 1921) was primarily aimed at 

estimating the economic impact caused by this insect and the methods to control 

it. Mackenzie estimated an annual financial loss of Rs.l.5 lakhs for plantations in

Burma and recommended that providing nesting boxes for birds that feed on 

defoliator larvae will help control the problem. 

Attempts were made seventy years back by Beeson to map the outbreaks 

of defoliator at the same general location of the present study. Starting from 

August 1926, a special officer was put in charge to patrol the teak plantations at 

Nilambur and record the distribution and grade ofdefoliation (Beeson, 1928). The 

study showed that complete foliage loss due to the insect occurred during the 

months September and October. Since the observer did not confirm the presence 

of insect in the area, it is difficult to draw any conclusions as to the progression of 

outbreaks. Moreover the incidence ofteak defoliator and teak skeletonizer was not 

distinguished while preparing the maps making it difficult to understand which 

insect caused damage when. This study indicated that control ofthe insect during 

the epidemic phase would be difficult and hence attempt has to be made to prevent 

the shift from endemic to epidemic phase. Beeson highlighted the difficulty in 

timely detection of outbreaks and commented that attempts to control teak 

defoliator requires the same alertness, as that demanded by forest fires. 

In 1934, a set of silvicultural-cum-biological control measures was put 

forward (Beeson, 1934). They were: (a) sub-division of large blocks of pure teak 

(sub-division by means of pre-existing forest rather than of newly created stands 

or mixtures); (b) establishment of a varied flora under the teak canopy (at the 

outset by retention of coppice re-growth and miscellaneous seedlings rather than 

by artificial introduction of selected species at a later stage); (c) elimination of 

harmful plants (this category includes alternative food-plants of defoliator); (d) 

maintenance of an understorey in older stands (for its value as a shelter for 

beneficial animals and as obstacle to defoliators) (e) introduction of parasites and 

predators (after careful assessment ofthe defective factors oflocality). 

Establishment of a varied flora under the teak canopy to provide adequate 

breeding sites for natural enemies of the pest was tried in 1942-43 (Khan et al., 

1944). Two plantations nearly two miles apart with differing levels of 

6

undergrowth were compared with respect to the incidence of defoliation and 

presence of parasites. However, the experimental set up did not yield reliable 

conclusions. Eventhough the role ofparasites was not empirically proved, faith in 

Beeson's recommendations persisted for a long period. But none of the 

recommendation were put to practice due to various reasons (Nair et al., 1997) 

except for an order issued in the then Madras state prohibiting the cutting away of 

undergrowth in teak plantations (Kadambi, 1951). 

Intensive observations were made in June 1950 at the Nilambur teak 

plantations to identify the causes of defoliator outbreaks (Kadambi, 1951). These 

observations brought out the fact that some trees escaped defoliation amidst a 

completely defoliated stand. Based on observation on the intensity of defoliation, 

it was suggested that the presence of tender foliage at the time of larval 

appearance was the factor that predisposed trees to defoliation. Research on how 

some trees escaped defoliation was also recommended. 

One of the suggestions put forward by Kadambi in 1951 was to test the 

resistance of trees that were found escaped amidst a defoliated stand. This was 

attempted in a study started in 1983 in Kerala (Nair et al., 1997). Trees, which 

escaped defoliation during one year, were observed in the next year. Many of 

these trees were found infested and grafting from ten trees, which had escaped 

defoliation under natural conditions, were readily attacked when exposed 

artificially to the insect. This meant that there was no genetic resistance to the 

pest. A comparison of resistance in the different clones at Nilambur and Arippa 

orchards indicated that none were resistant. Since the clones were all from Kerala, 

it was concluded that search for resistance may be continued using clones from 

other parts of India and abroad. Another study using twenty different clones 

(Ahmad,1987) collected from southern parts of India was done in 1987. One 

among the ten clones from Tamil Nadu (Top slip) showed the highest resistance to 

defoliator attack and another clone from Kerala (Karulai) showed the highest 

growth increment. The study proposed intraspecific crosses between these two 

clones for further improvement towards pest resistance and higher yield. 

7

In a two-year light trap study at Jabalpur in 1978 and 1979 (Vaisharnpayan 

et al.,1983), collection of teak: defoliator moths was restricted to July, August and 

September. Two explanations were put forward: migration ofmoths and diapause. 

Although the importance of biological control agents was highly 

emphasized during the early period, aerial spraying of chemical pesticides was 

done in 1965 (Basu-Chowdhury, 1971) and 1978 (Singh et al.,1978). The first 

spraying was on an experimental basis in an area of 76 ha at Konni teak: 

plantations in Kerala. The second spraying was done at Barnawapara plantations 

in Madhya Pradesh. In the second spray application, very few larvae survived in 

the sprayed plots as compared to the untreated controls. Although it was claimed 

that there was no adverse effect on wildlife including birds, the facts remain that 

80 1. Malthion, 75 I. Fenitrothion and 260 kg. Carbaryl were deposited over an 

area of460 ha. 

Argument against aerial spraying of chemicals was put forward (Nair, 

1980) based on three major reasons: (a) a realistic estimate of loss due to 

defoliator attack is not arrived at to calculate the cost-benefit ratio of aerial 

spraying, (b) environmental hazards and (c) adverse impact on natural enemies of 

the pest. Large-scale application of chemical pesticides against the teak: defoliator 

has not been reported except in nurseries and private sector plantations, in recent 

years. 

An attempt was made during the period 1979 to 1982 to answer the long 

standing question of economic impact of the teak: defoliator (Nair et al., 1996). 

Experimental plots in a four year old plantation were given selective protection 

against one or both ofthe two major defoliators or left unprotected for a period of 

five years. Measurements of trees at the end of the experimental period showed 

that the annual increment loss is 3 m 3 per ha in 4-8 year old plantations at 64% 

stocking. Projections based on this estimate indicated that protected plantations 

8

could yield the same volume of wood in 26 years as unprotected plantations 

would yield in 60 years, provided other necessary inputs are given. 

Evidences for migration of the defoliator moths were independently 

brought out by two groups ofresearchers during the later part of 1980's (Nair and 

Sudheendrakumar, 1986; Vaishampayan et al., 1987). The first study based on 

survey of defoliation along the Western Ghats and detailed observation of 

infestation characteristics at Peechi and Nilambur in Kerala proposed a model for 

the population dynamics of teak defoliator with short-range migration of moth 

populations. The second study relied on eight-year light-trap data from Jabalpur 

and showed a close link between defoliator outbreaks and the arrival ofmonsoon. 

It suggested that Kerala situated at the extreme southwest part of the country is a 

centre oforigin ofactivity ofHpuera from where moths migrate northward along 

with the progression ofsouthwest monsoon. 

A synthesis of information on the population dynamics of teak defoliator 

appeared in 1988 (Nair, 1988). It dispelled the notion that diapause occurs some 

time during the life history of the insect. Instead, it placed migration as the cause 

ofabsence ofdefoliator activity during part of the year. In almost the same way as 

Kadambi suggested in 1951, it emphasized the relation between presence oftender 

foliage and the susceptibility to defoliator incidence. It was also brought out that 

defoliator incidence is not associated with stand and site conditions of teak which 

means that outbreaks cannot be prevented by increasing the stand vigor through 

silvicultural management practices. 

A renewed interest in the role of biological control agents in combating the 

defoliator attack was seen during the past decade. Of particular importance is the 

nuclear polyhedrosis virus that was isolated from defoliator larvae 

(Sudheendrakumar et al., 1988). In the same year, observations were made on the 

bird predators of defoliator larvae which showed that 58 species of birds were 

feeding on defoliator larvae during the months of March, June and July (Zacharias 

and Mohandas, 1990). Studies on the parasitoids of teak defoliator at Nilambur 

9

during 1983-94 and 1987-89 recorded 15 species- seven from larvae and eight 

from pupae (Nair et al., 1995). Effectiveness of the dueteromycetous fungi, 

Beuveria bassiana (Bals.) Vuill, in causing mortality to the defoliator larvae was 

studied in 1993 (Rajak et at.,1993). It showed that the early larval instars were 

more prone to fungal infection. It is curious to note that a sixth larval instar of 

H.puera was used in this study while none of the earlier or later studies indicates 

the presence ofthe same. 

In an attempt to understand the spatial distribution of defoliator outbreaks 

Nair and Mohanadas (1996) kept the road-side plantations at Aravallikavu, 

Valluvasseri, Karulai and Kariem-Muriem at Nilambur under observation during 

the pre-outbreak season in 1987. The study showed that the first noticeable event 

in the chain of events leading to wide spread outbreak of defoliator is the sudden 

occurrence of fairly high-density, tree-top infestations in small, discrete patches 

covering 0.5 to 1.5 ha. These infestations were proposed to be the transitional 

stage between an endemic population and an epidemic and were designated as 

epicentres from where wide spread outbreaks originate. 

10

3.1. INTRODUCTION 

CHAPTERIII 

GENERAL METHODS 

This chapter summarizes the general methods used in the study; additional, 

specific details are described in the respective chapters. The work involved three 

major types of investigations- study of spatial distribution of defoliator outbreaks, 

monitoring of moth populations using light-trap and field observations on moth 

behaviour. 

All investigations were carried out in teak plantations at Nilambur, in north 

Kerala (Fig.3.1.). Specific methods used for monitoring moth populations are 

described in Chapter 6, and for studying the field behaviour ofmoths in Chapter 7. 

3.2. THE STUDY AREA 

The study area is located between Latitudes 11°10' N and 11°25' N and 

Longitudes 76°10' E and 76°25' E, and fall within Nilambur North and Nilambur 

South Forest Divisions. The teak plantations cover an area of about 8516 ha 

spread out in a geographical area of 25,750 ha (Fig.5.1, Chapter 5). 

The spatial distribution of outbreaks was studied at two spatial scales- in a 

continuous block of about 1000 ha of plantations at Kariem-Muriem over a three 

year period and in the entire teak plantations at Nilambur covering over 8500 ha., 

over one year. 

The Kariem Muriem teak plantation is located in the Vazhikkadavu Forest 

Range ofNilambur North Forest Division, between latitudes 11°22.7' and 11°25.7' 

and longitudes 76°16.44' and 76°18.47'. This area, located 16 km from Kerala 

Forest Research Institute (KFRI) Subcentre at Nilambur was chosen for detailed

area of 50 ha. A group of ten grids referred to above formed a block and was 

under observation of a single individual. Two individuals trained to identify and 

report defoliator outbreaks were deployed in the area to assist in the study. Each 

ofthese observers was asked to complete one round ofobservation within a period 

of 15 days. 

Within each grid, the level oftender foliage and the presence or absence of 

defoliator outbreaks was observed by criss-cross perambulation. However, this 

method did not permit detection of very low populations of the insect, which 

required intensive search. Only populations, which caused visual defoliation ofthe 

tree, were detected. 

Weekly visits were made to the plantation to verify the reports from 

observers. In addition, whenever the observers reported an infestation, the site was 

personally visited to gather information on (1) the date of egg laying and (2) the 

area infested. Two visits were made for this purpose, one at the beginning of the 

infestation to determine the date of egg laying and the other at the end of the 

infestation to determine the area infested. The following procedures were used. 

Determination ofthe date ofegg laying: 

Larval samples were brought from each ofthe infested sites and were reared in 

the laboratory until they moulted. Based on the date of moulting, the date of egg 

laying was arrived at by back-calculation based on the time span needed for each 

previous larval instars (preoviposition period- 2 days, egg- 1 day, Instars I to V 

2,2,3,3 & 3 days respectively and pupa- 5 d). 

Determination ofarea infested: 

This was done usually when the insect was in the pupal stage because by that 

time the full damage to the tree would have occurred, making it easier to estimate 

the infested area. A sketch ofthe infested area was made based on landmarks like 

13

oads, streams, etc. on copies of plantation maps prepared earlier. The area was 

estimated using Geographic Information System (GIS) as described in Chapter 4. 

3.5. OBSERVATIONS IN THE ENTIRE NILAMBUR TEAK PLANTATIONS 

The study area at Nilambur was divided into 149 grids and 20 observers were 

employed to report defoliator incidence. The area under supervision ofeach ofthe 

observers was visited at least once every week to verify their observation. 

Whenever outbreak was reported, the date of egg laying at the site and the area 

under outbreak was determined as described above. 

14

CHAPTER IV 

STUDIES ON THE SPATIAL DISTRIBUTION OF OUTBREAKS 

IN KARIEM MURIEM TEAK PLANTATIONS AT NILAMBUR 


The teak defoliator outbreaks are characteristic in their sudden occurrence over 

large plantations. It has been observed that outbreaks are prevalent only during 

some part ofthe year (Beeson, 1941). Light-trap collections ofdefoliator moths at 

Jabalpur, Madhya Pradesh (Vaishampayan et al., 1983) showed that a large 

number of moths were collected all of a sudden in July, preceded by a period of 

nearly 6 months when no moths were collected. This suggested that the insect is 

not breeding locally. Either migration or diapause was thought to be influencing 

the population level. A later study in Kerala (Nair and Sudheendrakumar, 1986) 

showed that the insect is active continuously in the teak plantations eventhough 

the population density fluctuated over the period - large scale outbreaks occurred 

during April-July and a very low population comprising overlapping generations 

ofthe insect was present during the rest ofthe year. In a three year study based on 

sample plots (Mohanadas, 1995), it was inferred that several distinct phases were 

recognized in the population trend of teak defoliator. The first phase during 

February to April is characterized by small patch infestations. This is followed by 

heavy and widespread infestations. It was observed that in a given large area, a 

second outbreak might occur before the moths of the existing generation has 

emerged. In the third phase, the population density declines and infestations 

become erratic. Following a lull period, erratic infestations occur again in August, 

September, or October and subside. Following this, it was observed that until the 

first phase begins again next year, the population remains very low, almost 

undetectable. 

However, very little is known on the distribution of defoliator outbreaks 

in space. Maps of plantations showing defoliation prepared by Beeson (1928)

showed that outbreaks do not occur simultaneously in all places. Based on a 

ground survey conducted in roadside teak plantations along the Western Ghats in 

Kerala and part of Kamataka, Nair and Sudheendrakumar (1986) showed that 

moths emerging from an outbreak site moved at least 4 km before causing another 

outbreak. They suggested a short-range, gypsy-type of movement of H puera 

populations resulting in a south to north progression in the incidence of outbreaks 

in course oftime. A later study at Nilambur (Nair and Mohanadas, 1996), showed 

that the first outbreaks during an year occur in a few small patches, 0.5-1.5 ha in 

area, which are widely separated. It was suggested that these early patches serve 

as epicentres where the population builds up and spread to other areas. 

Except the above few studies, most studies on H puera populations were 

concerned with temporal changes in population. Study of the spatial distribution 

of outbreaks is important to: (1) understand the cause-effect relationship between 

previous and subsequent outbreaks, and (2) examine the spatial preference of 

defoliator outbreaks. Detailed investigation made into the spatial distribution of 

outbreaks in about 1000 ha ofteak plantations at Kariem - Muriem during a period 

of three years, are described and analyzed in this Chapter. The pattern of 

outbreaks and its relationship with the topography ofthe area was examined using 

Geographic Information System (GIS). It was examined whether populations of 

the insect noticed within the study area could cause the subsequent outbreaks in 

the area. 

4.2. METHODS 

GIS was used to map the sites of infestation and relevant site characteristics such 

as elevation and aspect within the study area and to make relevant analysis of 

data. GIS is a computer software that store, retrieve, transform, display, and 

analyze spatial data (Anonymous, 1995). Georeferenced data, such as insect 

densities, crop type, or soils can be incorporated in a GIS to produce map layers 

(Liebhold et al., 1993). A map layer, generally composed ofonly one type ofdata, 

thus has a theme. The GIS serves as a tool for analyzing interactions among and 

16

within these various spatially referenced data themes. The software used was 

ARC/INFO in a UNIX platform. The methods used to prepare the maps and the 

procedure ofanalysis are given below. 

4.2.1 Preparation ofmaps 

4.2.1.1 Outbreak maps 

Defoliator outbreaks were mapped in the scale 1:50,000 as described in Chapter 3. 

These maps were digitized in individual layers ofthe GIS database. To understand 

the frequency of defoliator outbreaks in different sites within the study area, all 

the defoliation maps of a particular year were overlaid to produce a composite 

map. Information on the number oftimes that a particular site was under outbreak 

was generated in the database ofthe composite map. This information was used to 

produce the outbreak frequency map. 

FigA.1. Map of Kariem Muriem showing the layout of 

grids. 

17

4.2.1.2. Elevation and aspect maps 

The contour lines were scanned into a separate layer. A digital elevation model 

(DEM) was developed using the elevation values (z values) pertaining to the 

contour lines. The DEM contains a closely gridded surface with a particular 

elevation value assigned to each grid. The aspect map was prepared from the 

DEM. Aspect identifies the down-slope direction of the maximum rate of change 

in value from each cell to its neighbours. (Aspect can be thought of as slope 

direction). Aspect is expressed in positive degrees from 0 to 360, measured 

clockwise from the north. The values ofthe output grid are the compass direction 

of the aspect. The aspect map was generated as per methods provided by 

ARCIINFO; the details are given in Appendix 1. 

4.2.2. Spatial autocorrelation analysis 

Spatial autocorrelation is a measure of the similarity of objects (outbreak patches 

in this case) within an area. It was used to measure the relationship of defoliation 

frequency values of grids and the distance between them. Two autocorrelation 

indices were used in the present study- Geary index and Moran index. The indices 

are measures of attribute similarities as a function of distance. Algorithms for the 

indices were provided by ARCIINFO (Anonymous, 1995) and are given in 

Appendix A. The interpretations of Geary and Moran indices are summed up in 

Table 4.1. 

Table 4.1. Interpretation ofGeary and Moran indices 

Geary (c) Moran (1) Interpretation 

Ol 1

4.2.3. Correlation between outbreak incidence and topographic features 

The correlation between the defoliation frequency map with the topographic 

layers of elevation and aspect was calculated. For each pair of layers the 

covariance was calculated using the formula provided by ARCIINFO 

(Anonymous, 1995). 

4.3. RESULTS 

4.3.1. Outbreak pattern, 1992 

The sequence of outbreaks during the year 1992 is given in Table 4.2 and Fig.I. 

Systematic observations at Kariem Muriem were started in June, but prior to this, 

in April, an outbreak was detected at a small patch. Since the area was not 

estimated, this outbreak was excluded from the spatial analysis. It is not known 

whether similar outbreaks occurred at other places before June. The first outbreak 

after the start ofthe study period occurred on 13 June at two distinct patches. An 

area of 1.8 ha was totally defoliated during this infestation. The subsequent 

outbreak on 7 July occurred at three distinct patches and extended to an area of 

97.8 ha. Both the above said infestations occurred in different grids. Egg-laying 

was restricted to the top level ofcanopy in both the instances. 

During the first fortnight of August, a new infestation was recorded in an 

area of 10.5 ha. There were two distinct patches quite close to the sites ofthe first 

outbreak. Two days later, a larger area of 111.3 ha was infested. There were two 

infestations in September. The first extended to an area of 35.8 ha while the 

second to an area of 131.8 ha. A major infestation covering an area of 169.6 ha 

occurred during the first fortnight of October. The last infestation during the year 

was on 14 October covering a total area of21.4 ha. 

19

Table 4.2. Sequence ofdefoliator outbreaks at Kariem Muriem during 1992. 

SI. Date ofegg- No. of rea unde Probable date of Whether the actual date of 

No. laying ofthe outbreak outbreak egg-laying ofthe egg-laying (Column 2) 

observed patches (ha) resultant progeny falls within the range of 

population (Fl generation) probable dates ofegglaying 

by a previous 

generation ofmoths 

- 18 April unknown small 8-13 May Unknown 

patch 

1 13 June 2 1.8 3-8 July No 

2 07 July 3 97.8 27July-01 August Yes 

3 03 August 2 10.5 23-28 August No 

4 06 August 4 111.3 26 August-O1 No 

September 

5 01 September 3 35.8 21-26 September Yes 

6 22 September 5 131.8 12-17 October Yes 

7 01 October 4 169.6 21-26 October No 

8 14 October 2 21.4 3-8 November Yes 

Total area infested (ha) 580.0 - - 

Thus, there were eight outbreaks during the year. Based on the date of egg-laying 

and the information on life span of the different life stages of the defoliator the 

probable date of start of Fl generation can be computed. It can be seen (Table 

4.1.) that the date of egg-laying which caused populations 2,5,6 and 8 overlaps 

with the start date of FI generation of a previous population. It is quite probable 

that the populations are the offsprings ofearlier populations. However, it is certain 

that populations 1,3,4, and 7 could not be caused by the offsprings of earlier 

populations. The populations that could have been caused by earlier populations 

comprise a total area of 286.8 ha out of 580 ha infested during the year. This 

means that nearly 50% of the infestations during the year 1992 could have been 

caused by the offspring ofearlier outbreaks during the year. 

20

4.3. 2. Outbreak pattern, 1993 

The sequence ofoutbreaks during the year 1993 is shown in Table 4.3. and Fig. 2. 

The first outbreak during the year 1993 occurred on 19 February at two distinct 

patches comprising a total area of 5.8 ha. One week later, a single patch of 

infestation was noticed at a different site. It was a small patch of 2.5 ha and had 

the typical tree top infestation. A much larger outbreak occurred on 20 March 

extending to the entire southern part of Kariem Muriem. The area under outbreak 

was 549.1 ha. The fourth outbreak, which extended to 235.8 ha started on 03 

April. It was confined to the northern part of Kariem Muriem. The next outbreak 

occurred on 18 April in a single patch covering an area of 290.3 ha. The sixth 

outbreak occurred on 15 May covering an area of 720 ha. The largest outbreak 

during the year occurred on 10 June extending to an area of 810.2 ha. Nearly one 

month later, the eighth outbreak occurred in an area of 52 ha. The last two 

outbreaks during the year occurred on 27 August and 1 September extending to an 

areaof36.1ha and 35.7 ha respectively. 


SI. Date of egg- No. of Area under Probable date ofegg- Whether the actual date 

No. laying ofthe outbreak outbreak laying ofthe resultant of egg-laying (Column 

observed patches (ha) progeny (Fl 2) falls within the range 

population generation) of probable dates ofegglaying 

by a previous 

generation ofmoths 

1 19 February 2 5.8 11-19 March No 

2 26 February 1 2.5 18-26 March No 

3 20 March 1 549.1 9-17 April Yes 

4 03 April 1 235.8 23 April- 01 May No 

5 18 April 1 290.3 8-16 May No 

6 15 May 1 720.0 4 - 12 June Yes 

7 ] 0 June 1 810.2 30 June - 8 July Yes 

8 7 July I 52.0 27 July - 4 August Yes 

9 27 August I 36.1 19 - 30 September No 

]0 I September I 35.7 24 Sep. - 02 Oct. No 


It can be seen that outbreaks at serial No.s 3, 6, 7, and 8 could be explained 

as caused by progenies of earlier populations. Outbreaks caused by these 

populations extended to an area of2131.3 ha (78%) Out ofthe total area of 2737.5 

ha infested during the year. 

22

g ID May h 7 July 

27 August J I September 

Fig.2. (contd) Sequence of defoiiator outbreaks in 1993. 

• 

24

4.3.3. Outbreak pattern, 1994 

Table 4.4 shows the sequence of outbreaks during the year 1994. The first 

outbreak was on 04 April which extended to an area of 12.2 ha (Fig.3). The 

second outbreak occurred at two different places on 12 May. The third outbreak 

was on 03 June in a single patch covering 75 ha. The last outbreak occurred about 

a week later on 12 June and extended to an area of435.3 ha. 


SI.N Date ofegg- No. of Area under Probable date Whether the actual date of 

laying ofthe outbreak outbreak ofegglaying of egg-laying (Column 2) fall 

observed patches (ha) the resultant within the range of probable 

population progeny (Fl dates ofegg-laying by a 

generation) previous generation ofmoths 

1 04 April 1 12.2 25-30 April No 

2 12 May 2 472.5 01-06 June No 

3 03 June 1 75.0 23-28 June Yes 

4 12 June 1 435.3 01-06 July No 


It can be seen that only the third population could have been caused by any 

of the previous populations. This population extended to an area of 75 ha out of 

995 ha infested during the year. Thus, the infestations that could be explained 

based on earlier populations comprised only 7% ofthe total area infested in 1994. 

Populations 1,2 and 3 did not cause any further outbreaks in the study area. 

4.3.4. Frequency ofoutbreaks in space 

The frequency map of defoliation was generated for each of the years (Fig. 4,5 

and 6). The area under each of the frequency class is given in Table 4.5. It may be 

seen that the outbreak frequency was higher during 1992 and 1993 compared to 

that of 1994. 

25

a 4 April 

(j 

a 

tJ 

(J 

• 

b 12 May 

c 3 June d 12 June 

Fig.3. Sequence ofleak defolialor outbreaks in 1994. 

26

_ 0 

_ 1 

0 2 

_ 3 

_ 4 

_ 5

_ 0 

_ 1 

0 2 

_ 3 

_ 4 

_ 5 

F1g.5. Frequency or deroliator outbreaks In 1993.

_ 0 

_ 1 

0 1 

_ 3

Table 4.5. Area under each ofthe outbreak frequency class during the years 1992- 

94. 

Frequency Area infested (ha) 

class 1992 1993 1994 

0 596.6 179.1 310.8 

1 286.6 523.9 359.7 

2 85.4 242.0 299.4 

3 11.6 35.9 12.2 

4 1.7 0.7 - 

5 0.2 0.5 - 

Total 385.5 803.0 671.3 

In 1992, the outbreaks were confined to small patches, which left more 

than half of the area uninfested. It can also be noticed that the sites of maximum 

infestation during all the three years were in close proximity to each other. 

Eventhough there were a large number of outbreaks in all the three years (8,6 and 

4 during 1992,1993 and 1994, respectively), there were still places where no 

infestation occurred. The results of spatial autocorrelation analysis are given in 

Table 4.6. 

Table 4.6. Spatial autocorrelation indices for the years 1992-94. 

Year Geary index Moran index 

1992 0.039960 0.95514 

1993 0.033901 0.95600 

1994 0.038924 0.95603 

In all the years, the Geary index had a value between zero and one and the 

Moran index was greater than zero. This shows that the defoliator outbreaks are 

regionalized, smooth and clustered (Table 4.1). The sites with the same outbreak 

frequency were adjacent to each other. 

4.3.5. Correlation between outbreak incidence and topographic features 

The elevation ofthe entire study area ranged from 35.9 m to 283.4 m. (Table 4.7.). 

30

Table 4.7. The relationship between outbreak frequency and elevation. 

Outbreak Elevation (m) 

frequency 1992 1993 1994 

0 97.1+44.1 132.1+44.3 118.3+53.4 

1 108.7+46.1 104.4+48.9 103.8+44.7 

2 138.5+55.3 81.2+28.9 87.0+35.8 

3 71.3+19.0 57.2+15.5 51.4+9.9 

4 68.7+6.2 58.4+ 1.1 - 

5 68.4+2.1 44.4+2.0 - 

The elevation ofsites with the highest frequency ofoutbreaks was between 

40 to 77 metres above sea level. During 1992, a positive correlation was found 

between the outbreak frequency and elevation. However, in 1993 and 1994, the 

correlation was found to be negative. This indicates that the elevation of a site 

may not be determining the susceptibility of a site to defoliation. The details of 

aspect for the frequency classes are given in Table 4.8. 

Table 4.8. The relationship between outbreak frequency and aspect. 

Frequency Aspect (degrees) 

class 1992 1993 1994 

0 171.1+86.9 180.1+89.3 187.8+88.3 

1 188.2±94.2 168.5±85.5 158.8+86.3 

2 224.3+77.5 204.1+92.0 197.5+87.7 

3 272.0+21.6 226.8+88.7 284.3+58.0 

4 256.6+13.6 296.6+14.7 - 

5 240.8+4.6 298.7+8.3 - 

It can be seen that there is an increase in the aspect value corresponding to 

an increase in outbreak frequency. It is indicated that the high frequency sites face 

predominantly to the west while the low frequency sites face to the south. Sites 

that were not infested during the three years faces predominantly to the south. It 

was found that the aspect values were positively correlated to outbreak frequency 

in all the three years. 

Table 4.9. shows the elevation and aspect of the first outbreaks. The first 

outbreaks are shown in landscape perspective in figures 17-19. 

31

Fig. 7. Location offirst outbreaks at Kariem Muriem in 1992 in landscape 

perspective. 

32

Fig. 8. Location of first outbreak at Kariem Muriem in 1993 in landscape 

perspective. 

Jl

fi g. 9. Location of first outbreak at Kariem Muriem in 1994 in landscape 

perspective. 

34

Table 4.9. Elevation and aspect ofthe first outbreak sites at Kariem Muriem in the 

years 1992-1994. 

Year Elevation Aspect 

(mts.) (degrees) 

1992 70.8 ± 3.1 284 ± 41 

1993 75.7 ± 19.2 247 ± 73 

1994 51.9±10.1 288 ± 50 

It can be seen that the site of occurrence of first outbreak in 1994 was an area that 

had the first outbreaks in 1992 and 1993. Small patches occurred outside this area 

in 1992 and 1993. Generally, these sites had a mean elevation ranging from 50-75 

metres and mean aspect ranging from 247-288 degrees. Ifthe relationship between 

susceptibility to defoliation and topography were known, it would greatly reduce 

the area to be monitored for identifying initial outbreaks. Observations at other 

teak growing areas are needed to generalize these findings. 

4.4. DISCUSSION 

The pattern of defoliator incidence in all the three years indicated that the 

outbreaks were not randomly distributed in space. It was observed that in all the 

three years, sites having the highest outbreak frequency value were surrounded by 

an area with the next lower frequency value. Spatial autocorrelation indices of the 

outbreak frequency maps indicated that the high frequency sites occur in a 

clustered manner. This indicates that some sites are more prone to defoliator 

attack. This is similar to the spatial pattern exhibited by the Gypsy moth 

(Lymantria dispar L.) defoliation (Liebhold and Elkinton, 1989). 

The temporal sequence of outbreaks given in Tables 4.2, 4.3, and 4.4 shows 

that the outbreaks which occur during any year are not always caused by 

generations of the insect breeding in the same area. Out of the total of eight 

outbreaks in 1992, only those on 7 July, 6 August, 22 September, and 14 October 

could have been caused by moths emerging from the same area. The only possible 

cause for the other four outbreaks is the immigration ofmoths into the study area. 

35

In 1993, out of the total often outbreaks only four (which occurred on 20 March, 

15 May, 10 June, and 7 July) could have been caused by moths emerging from the 

same area. Immigration has occurred on 19 February, 26 February 3 and 18 April, 

27 August and 1 September. In 1994, only one outbreak that occurred on 3 June 

could have been caused by moths emerging from one of the previous outbreak 

sites in the area. During the year, immigration of moths caused outbreaks on 4 

April, 12 May and 12 June. 

A further proofofthe movement ofmoths is the fact that moths emerged from 

many of the outbreak sites did not lay eggs within the study area. In 1992, of the 

eight distinct outbreaks only four could have caused further outbreaks within 

Kariem Muriem. In 1993, only 4 out of 10 and in 1994 only 1 out of4 could have 

caused similar outbreaks. The moths emerged from the other outbreaks should 

have emigrated from the study area. 

The indication that a particular outbreak was caused by moths emerging from 

an earlier population in the same site is solely based on temporal correspondence. 

When emergence of moths from a particular population was found to coincide 

with the egg-laying which initiate another, it is assumed that the first population 

caused the second. However, if the moths that emerged move out of the study 

area, and a new group of moths arrived and laid eggs, they could not be 

distinguished. This indicates that there are chances that migration may be more 

frequent than can be proved as above. 

The following conclusions can be drawn from the present study: 

a) Occurrence of defoliation is not randomly distributed in space. The sites with 

higher outbreak incidence occur close together and some areas are more prone 

to defoliator attack than others. 

b) There is no significant correlation between the frequency of outbreaks and the 

elevation of the site. However, there was significant relationship between 

36

aspect and outbreak frequency. The high frequency sites faced predominantly 

to the west while the low frequency sites faced either south or south 

southwest. 

c) Migration of moths plays an important role in the local population dynamics 

of the insect. Majority of outbreaks have been proved to be caused by moths 

migrating into the area. 

The fact that most of the outbreaks could not be explained as caused by the 

progeny of earlier populations could be because the area was too small for a 

spatial study. Immigration of moths from a larger surrounding area may explain 

the origin of presently unexplainable populations. This inference was apparent 

from the 1992 results were only 4 out of 8 populations could be explained based 

on earlier populations at Kariem Muriem. Observations covering all the teak 

plantations at Nilambur were made in the year 1993 to understand if outbreaks 

occurring in a larger area could be caused entirely by progenies of earlier 

populations in the area; the results are presented in the next Chapter. 

37

CHAPTER V 

STUDIES ON THE SPATIAL DISTRIBUTION OF OUTBREAKS 

IN THE ENTIRE TEAK PLANTATIONS AT NILAMBUR 


In 1992, the outbreak pattern in the Kariem-muriem teak plantation showed 

that out of eight outbreaks, only four occurred when local moth populations 

were present. The rest of the outbreaks could only be explained by 

immigration of moths from elsewhere into this area. Therefore in 1993, an 

attempt was made to understand the dynamics of outbreaks in a larger spatial 

scale. 

Using the same methodology adopted in 1992 for Kariem- Muriem, the 

entire teak plantations at Nilambur were observed for incidence of defoliation 

in 1993. All the outbreaks that occurred in the 8516 ha of plantation were 

mapped and analyzed using GIS. The objectives were to characterize the 

spatial and temporal pattern of defoliator outbreaks in a large plantation area, 

and to study the influence of scale in our understanding of the population 

dynamics ofthe insect. 

5.2 METHODS 

As noted in Chapter 3, the teak plantations of Nilambur extend to an 

area of 8516 ha spread out in a geographical area of 25,750 ha. The major 

continuous blocks of plantations are located at Nedumgayam, Sankarankode, 

Ezhuthukal, Poolakkappara, Nellikkutha, Kariem- Muriem, and Edakkode 

(Fig. 5.1). The area was divided into 149 grids and fortnightly observations 

were made in each of the grids. Trained individuals were employed to assist in 

identifying outbreaks. Whenever an outbreak was detected, live insects were 

collected to identify the developmental stage. During the pupal period of the 

population, the area was visited to map the outbreak area in the plantation

map. The methodology adopted for data collection and analysis were similar 

to that described in Chapter 4. 

5.3 RESULTS 

The temporal sequence of defoliator outbreaks at Nilambur is given in 

Table 5.1. The first outbreak of the year occurred at Kariem-Muriem on 19 

February (Fig 5.1). There were two distinct patches of infestation, separated 

by a distance of about 3 km, one covering an area of 12.8 ha and the other, 1.7 

ha. 

Table 5.1. Chronology ofdefoliator outbreaks at Nilambur in the year 1993. 

Serial Date ofegglaying Infested No. of Expected egglaying 

No. area (ha) patches period ofprogeny 

1 19 February 14.3 2 11·19 March 

2 26 February 10.0 1 18-26 March 

3 17 March 38.8 1 06-14 April 

4 20 March 512.0 1 09-17 April 

5 21 March 1.7 I 10-18 April 

6 26 March 0.12 1 18-23 April 

7 03 April 254.4 3 23 April-Ol May 

8 07-20 April 934.4 24 27 April-IS May 

9 23 April 11.9 1 13·21 May 

10 25-26 April 18.1 4 15-24 May 

11 28 April 1.5 2 18-26 May 

12 05 May 114.7 3 25 May-02 June 

13 08-30 May 2498.9 47 28 May-27 June 

14 02-16 June 2531.5 67 22 June-18 July 

15 28 June 4.25 L 21-30 July 

16 01 July 22.4 1 24 July-2 August 

17 04-11 July 208.6 16 27 July-12 August 

18 14 July 2.65 1 06-15 August 

19 18 July 0.5 1 10-19 August 

20 27 August 40.6 1 19·30 September 

21 01 September 35.7 3 24 Sept.-02 act. 

22 03-06 September 1.3 4 26 Sept.-08 act. 

23 08-09 September 1.6 3 01-11 October 

Before this, there was no visible evidence of presence of larvae although 

occasionally stray larvae could be located on careful search. General 

39

observations had indicated that in most areas, leaf fall was completed by the 

end of January and new flushes had started. The infestation, which occurred 

on 19 February, was dense and concentrated on the treetops. The total area 

under outbreak was 14.3 ha. This outbreak could only be caused by a large 

number of moths. It is obvious that such a large population ofmoths could not 

have originated from the teak plantations of Nilambur all of a sudden, unless 

stray moths which were spread throughout the plantations got concentrated 

through some unexplained phenomenon. 

The second outbreak occurred near one of the first outbreak patches on 

26 February (Fig 5.2) covering an area of 10 ha. Since only seven days had 

elapsed between the first and second outbreaks, it is obvious that the second 

outbreak was not the progeny of the first. The 7-day gap between the two 

outbreaks also suggests that the second outbreak could not have been caused 

by the same group of moths, which caused the first outbreak because the 

normal egg laying period is only 5 days. In view of the above facts, the origin 

of the second infestation cannot be explained except in the same manner as the 

first. 

The third outbreak occurred on 17 March in a different plantation 

(Fig.5.3) and covered an area of 38.8 ha. In theory, the moths which caused 

this infestation could be the progeny of the first population (moth population 

at Serial No.l, Table 5.1). 

The fourth outbreak occurred on 20 March and covered an area of 512 

ha in the general location ofthe first two outbreaks (Fig.5.4). This was the first 

large-scale infestation wherein over 500 ha were infested on a single day. This 

infestation could be attributed to the progeny of the moth population at Serial 

No.2; Table 5.1. 

The fifth infestation was noticed on 21 March and covered an area of 

1.7 ha in a different location (Fig.5.S). In theory, this could also be attributed 

to the progeny of population No.2 which caused the large infestation on the 

previous day a different location. However, the very small area of this 

outbreak has some similarity to the first two infestations ofunexplained origin. 

40

The sixth outbreak occurred on 26 March in a new area (Fig.5.6) and 

covered only 0.12 ha. In theory, this can also be attributed to the progeny of 

population No.2, but the very small area ofinfestation and the gap between the 

current and the previous infestation indicates that this is unlikely. As in the 

previous infestation, the similarity of the infestation to the first two of 

unexplained origin is striking. 

The seventh outbreak occurred on 3 April in the same general area 

where the first two outbreaks occurred and covered a substantial area of about 

254.4 ha (Fig.5.7). This was the second major infestation of the year in terms 

of area coverage. The origin of this infestation could not be attributed to any 

ofthe previous populations within Nilambur (Table 5.1). 

In the following period, a very large area (934 ha) was infested from 7 

to 20 April. It was not possible to distinguish the area infested on each day 

within this interval (Fig.5.8). There were 24 distinct patches of infestation, 

which included some very small patches similar to the initial infestation 

patches. All these infestations can be explained, in theory, as caused by the 

progeny ofprevious populations (Table 5.1). 

All the subsequent infestations which occurred from 23April to 18 July 

(serial No. 9 to 19 in Table 5.1and Figures 5.9 - 5.19) were also attributable 

(in theory) to the progeny of previous populations within the area. However, 

the possibility of immigration or local concentration cannot be ruled out. 

These infestations (Serial No. 9 to 19) constitute the major part of the 

infestations covering an area of5415 ha (88.42%) out ofthe total infested area 

of7180 ha. 

The last four infestations of the year from 27 August to 9 September 

(serial No. 20,21,22 and 23 in Table 5.1 and Figures 5.20-5.23) were not 

attributable to previous populations, in the same way as the first few 

infestations. This also indicates either immigration of moths from outside the 

study area or aggregation ofstray moths from within the area. 

41

Fig. 5.24 is an overlay of areas infested in all the outbreaks during 

1993 ; green areas were not infested at all, each infestation frequency is 

indicated by a different colour. It may be seen that large areas of teak 

plantations were not infested during the year. At the same time, some areas 

were repeatedly infested, upto 5 times. It is clear that not all plantations were 

equally prone to defoliator outbreak. Plantations in two locations viz. Kariem 

Muriem, and Sankarangode - Nedumgayam - Kanjirakkadavu were the most 

heavily infested in 1993. 

Spatial autocorrelation indices were 0.08638 (Geary) and 

0.4051(Moran), which indicated that the outbreak frequency classes were 

regionalized, smooth, and clustered. 

5.4 DISCUSSION 

If we analyze the first seven outbreaks, which occurred on single days 

(Table 5.1), it can be seen that three of them (1sI 2 nd and 7 th ), cannot be 

attributed to the progeny of pre-existing populations within the study area 

consisting of about 8,500 ha of teak plantations. This clearly shows either 

immigration of moths from other areas or aggregation of moths dispersed 

throughout the teak plantations in Nilambur. The Sth and 6 th infestations, 

covering very small areas may also fall in this category, as indicated above. 

As noted earlier, this study in a large area was made necessary 

because of the inability to explain all outbreaks which occurred in 1992 in a 

1000 ha plantation (Kariem Muriem)as caused by progenies of earlier 

population during the year. During 1993, the outbreaks which occurred at 

Kariem Muriem plantations defoliated an area of 2737.S ha out of which 

2131.3 ha (about 78%) could be explained as caused by progenies of 

population which were present in the same area. Out of ten outbreaks during 

the year, six outbreaks (which occurred on 19 and 26 February, 3 and 18 

April, 27 August and 1 September) could not be explained in this manner. 

When the entire teak plantation at Nilambur was studied, it was seen that the 

outbreak which occurred on 18 April (at an area of 290.3 ha) could be 

42

explained as caused by progenies of earlier populations in the entire Nilambur 

area. Thus percentage area infested by populations which could be caused by 

earlier populations increased to 88% when the populations in the entire 

Nilamur area was considered. 

A comparison of temporal sequence of outbreaks which was observed 

at Kariem-Muriem and at all the plantations at Nilarnbur, including the 

Kariern-Muriem plantations, shows that when the larger area is considered, the 

sequence of outbreaks become more explainable. The percentage of outbreaks 

that could be attributed to local populations was computed for this purpose. 

Only about 33% (2 out of 6) of the outbreaks could be attributed to local 

populations when only Kariem-Muriem area was considered. Nevertheless, 

about 70% (16 out of 23) of outbreaks could be explained when all the 

plantations were taken together. This indicates that the chain of outbreaks that 

occurred during a year is more self-sustained when viewed in a larger area. 

Short-range migration of moths within Nilambur can explain this 

phenomenon. When only Kariem-Muriem is considered, the effect of 

emigration or immigration will lead to unexplainable outbreaks within 

Kariem-Muriem. However, when all the plantations at Nilambur are 

considered the effect ofshort-range movement is only spatial displacement but 

temporally the outbreaks remain explainable. 

The salient features of the spatial and temporal pattern of outbreaks 

can be summarized as follows: 

(a) Temporal pattern 

i. The period of defoliator outbreaks at Nilambur extends 

from February to September. 

11. The initial outbreaks that occurred in February and March 

in 1993 could theoretically cause a sequence of outbreaks 

which comprised 78% of the total area under outbreak 

during the year. 

111. The outbreaks which occurred during the later part of the 

year (August and September in 1993) could not be caused 

43

(b) Spatial pattern 

by the initial populations which indicate that there is a 

break in the sequence ofoutbreaks. 

i. All the infestations occurred in discrete patches in spite of 

the existence ofcontiguous infestable plantations. 

ii. Not all plantations were equally infested. While some 

plantations remained uninfested, some were infested up to 

five times during the year. 

iii. Sites with high frequency of outbreaks were clustered 

together indicating that outbreak incidence is not a random 

phenomenon. 

This study suggests that controlling the initial populations of the 

insect so as to prevent population build up leading to large scale outbreaks is a 

valid proposition, since it has shown that the initial populations could 

theoretically cause nearly 70% of the outbreaks that occur during the year. 

This can be confirmed by controlling the initial populations and then 

monitoring the entire plantations. The origin of initial populations during the 

year, those during the final phase of outbreaks and a single outbreak in 

between is still uncertain. Two possible explanations are (a) aggregation of 

stray moths from the same locality, and (b) long-range immigration of moths. 

Monitoring of defoliator outbreaks in wider geographic regions can give 

indication whether long-range movement of moths causes the presently 

unexplained outbreaks. 

44

Fig.5.3. Map ofNilambur teak plantations showing the locations ofthe 

third outbreak in the year 1993 on 17 March. 

47

o b 

Fig.SA. Map ofNilambur teak plantations showing the locations of the 

fourth outbreak in the year 1993 on 20 March. 

48


fifth outbreak in the year 1993 on 21 March. 

49


sixth outbreak in the year 1993 on 26 March. 

50


seventh outbreak in the year 1993 on 3 April. 

51


eighth outbreak in the year 1993 during 7-20 April. 

52


ninth outbreak in the year 1993 on 23 April. 

53

Fig.5.14. Map of NilambUT teak plantations showing the locations ofthe 

forteenth outbreak Inthe year 1993 during 2·16 June. 

S8

Fig.5 .15. Map ofNilambur teak plantations showing the locations ofthe 

fifteenth outbreak in the year 1993 on 28 June. 

59


sixteenth outbreak in the year 1993 on 1 July. 

60

Fig.5.l9. Map ofNilambur teak plantations showing the locations ofthe 

nineteenth outbreak in the year 1993 on 18 July. 

63

Fig.S.20. Map ofNilambur teak plantations showing the locations ofthe 

twentieth outbreak in the year 1993 on 27 August. 

64

Fig.5 .21. Map ofNilambur teak plantations showing the locations ofthe 

twenty-first outbreak in the year 1993 on 1 September. 

65


twenty-second outbreak in the year 1993 during 3-6 September. 

66

'------------------ ------- --- ----------------------_._..__.----_._._-------' 

Fig.S.23. Map ofNilambur teak plantations showing the locations ofthe 

twenty-third outbreak in the year 1993 during 8-9 September. 

67


CHAPTER VI 

FIELD OBSERVATIONS ON MOTH BEHAVIOUR 

This chapter summarizes the field observations made on aggregation, 

mating, oviposition, and flight behaviour of the teak defoliator moth. Laboratory 

studies made in the past have given information on the oviposition and 

reproductive behaviour (Sudheendrakumar, 1994). However, very few studies 

have been reported on the behaviour ofmoths under natural conditions. 

Sudheendrakumar (1994) reported that under laboratory conditions, moths 

emerged from field collected pupae attained sexual maturity within 2 days while 

those from pupae reared in the laboratory took 3 days. Mating occurred 

predominantly in the second half of the scotophase, between 01.00 to 05.00 h. 

Duration of copulation was found to be ranging from 50-220 min. After mating, 

egg-laying started within 18-24 hours and continued up to a maximum period of 

11 days. The fecundity ranged from 287 to 606. Egg-laying was continuous in 

most ofthe (75%) observed moths. 

Relatively no information is available on the flight activity of the moth. 

However, short-range migration of the moths has been postulated to explain the 

observed spatial distribution of outbreaks (Nair and Sudheendrakumar, 1986). 

Aggregation of newly emerged moths has also been reported ( Nair, 1988). The 

fact that outbreak populations of the insect consist predominantly of a single 

instar had suggested that during outbreaks, egg- laying occurs at a site on a single 

day. The admixture of two instars was attributed to the difference in the 

developmental time between individuals.

6.2. METHODS 

In this study, post emergence behaviour ofmoths was studied at the site of 

emergence by establishing a floor-less cage within a teak plantation, when in 

March 1993, nearly 92 ha of teak plantations at Kariem Muriem was under 

infestation. A cage (3m x 2m x 2m) made of nylon net was established at a 

suitable site in the plantation in the first week ofApril when the insect population 

had reached the pupal stage. The ground within the cage was cleared of fallen 

leaves. Pupae were collected from the nearby area and were sexed. A total of 160 

pupae (100 female and 60 male) were placed on the floor inside the cage along 

with fallen teak leaves so as to provide a relatively natural microenvironment for 

the pupae. Once the moths started emerging, diluted honey was provided on 

sponge as food. At hourly interval, observations were made on the number of 

moths emerged and their behaviour. During night, a dim, red light was used to 

make the observations. Observations were continued for a period of three days. A 

trained field observer was employed to take observations from the cage when 

simultaneous observations had to be made in the field. Some observations were 

also made on the behaviour ofmoths emerging in the field, outside the cage. 

The flight behaviour of moths was observed at Kariem-Muriem during 

1993. Over a period ofone month period immediately following the emergence of 

moths from an infestation which occurred on 20 th March, observations were made 

on the movement of moths in the field. To assess the sex ratio, collections of 

moths were made from aggregations that were found on ground or in flight. Since 

it is known that mated females start to lay eggs within two days 

(Sudheendrakumar, 1994), the females collected were individually reared for two 

days to know whether they laid eggs. Absence of egg-laying was taken as an 

indication ofabsence ofmating. Whenever counts ofmoths in flight were taken at 

different sites at the same time, trained observers were deployed for the work. 

70

Observations on the oviposition behaviour ofmoths under field conditions 

were made at Valluvassery in Nilambur in a three-year-old teak plantation. 

During one of the routine defo1iator population monitoring exercises in 1993, a 

large number of moths were found hovering over the young trees at dusk. Careful 

observations revealed that the moths were laying eggs on the foliage. Next 

morning, 56 leaf pairs were marked in the plantation and the egg count was taken 

for each of them. Counting of eggs was repeated on the second and third day in 

the marked leaves. 

6.3. RESULTS 

6.3.1. Post-emergence behaviour 

Observations made within and outside the field cage established at Kariem 

Muriem showed the following: 

1. Freshly emerged moths were first found in the field at 18.00 h. Seven moths 

were collected out of which four were females. The first emergence of moths in 

the cage 

15 

., / 

1 10 

.... 0 

5 I 

cl 

z 

0 

r 

0 6 12 18 24 6 12 18 24 6 

Time of the day 

Fig.6.1. Graph showing the emergence ofmoths in relation to the time ofthe day. 

71

was observed at around 06.15 h on the next day (Fig.6.1.). Four moths were found 

of which two were females. One more moth emerged at 12.00 h. No emergence 

was noticed until 07.00 hrs on the next day. Ten moths were found emerged at 

07.00 h. At 09.00 h, 13 moths had emerged. No further emergence was noticed 

during the next day. 

2. The moths that emerged in the cage spent nearly I hour before they were able 

to fly. 

3. Feeding commenced after about 5 hours from the time ofemergence. 

4. A single pair of moths was found to start mating behaviour at 01.40 h. The 

duration ofcopulation was found to be 200min. 

5. Before mating, the females exhibited characteristic calling behaviour (lifting of 

wings, curving of abdomen and protrusion and retraction of terminal abdominal 

segments). 

6.0bservations outside the cage revealed that during the time of emergence of 

moths from a densely populated site, birds flock in and feed on the newly 

emerged moths while they are unable to move by flight. 

6.3.2. Aggregation 

Finding moth aggregations depended on chance. Since systematic 

observations are difficult, only limited data could be generated. The first 

aggregation of moths was found on 16 April 1993 on the top of a hillock at 

Thannikkadavu in Kariem Muriem. This area was under outbreak; egg-laying had 

occurred during the period 9-11 April. The moths were found on the undergrowth 

and moved only when disturbed. The group consisted ofboth freshly emerged and 

72

old moths of both the sexes. This indicates that moths emerged on different days 

can be present in a single aggregation. 

On the next day (17 April 1993), search was made for moth aggregations 

within the Kariem Muriem area. Aggregation was found on top of two hillocks 

Ambalakkunnu- in the middle of the Kariem Muriem and Enpathukunnu-about 1 

km north of Ambalakkunnu. Collections ofmoths were made both in the morning 

(4 males and 4 females) and evening (10 females and 3 males) from an area of25 

sq m at Ambalakkunnu. At both the places, fresh and old moths were found 

together. The females collected in the morning and evening laid eggs during night 

on the same day. No aggregations could be detected on the low-lying areas within 

the plantation. 

6.3.3. Flight behaviour and dispersal 

6.3.3.1. Movement ofmoths outside the teak plantation 

Directional flight by a group of defoliator moths was observed on 5 April 

1993 at Kariem Muriem. Moths were found to cross a paddy field and move into 

the teak plantation bordering it. The flight was at a height less than 5 m above 

ground level. A steady stream of moths was found moving in the same direction. 

This movement was detected at 15.45 h. Three hundred moths could be counted 

from a single observation point until the movement ceased at 16.40 h. It was 

observed that the movement of moths occurred in a wide area. Next day, moths 

were observed to move as on the previous day but towards the opposite direction 

(towards south) during the period 16.00 -19.30 h. Observations revealed that the 

flight path was nearly 200 m. wide. On the third day, movement of moths was 

detected at dusk towards south and away from the plantation. A total of 650 

moths were counted from two observation points. 

73

Two moths (one male and one female) were collected when in flight on 6 

April 1993. The female laid eggs on 8 April. Moths collected when in flight on 7 

April were predominantly females (28 females and 1 male). The females laid eggs 

on 8 April. 

6.3.3.2. Movement ofmoths within teak plantations 

The number of moths found in flight and the direction of their movement 

as recorded in observations made between 18.45 hr and 19.30 hr over a period of 

five days from 16 to 20 April are shown in Fig.6.2. 

s 

19 April N 

E E 

18.45-19.30 S 18.45-19.30 S 

20 April N 

18.45-19.30 

Fig.6.2. Number and direction ofmoths found in flight at Ambalakkunnu, Kariem 

Muriem. 

On one of the days, observations were also made between 05.45 hr and 

06.30 hr. An average of 1400 moths were observed in flight on each evening. 

Uni-directional flight was observed on two occasssions -16 th dusk and 18 th dawn. 

On other days, the movement was to different directions indicating diffusion of 

moths in the plantation. 

E 

74

6.3.4. Oviposition 

Egg-laying was found to start at dusk (6.55 p.m. to 7.10 p.m.) on all the 

three days observed. The females hover over the shoot for a while and settle on 

the leaves. Eventhough some moths settled on the older leaves, they moved on to 

the lower surface of nearby tender foliage. The female moth walks on the leaf 

with the tip of the abdomen touching the leaf surface. The moth lays a single egg 

at a time close to the veins of the leaf. Moths were found to oviposit on leaves 

which was oviposited earlier. 

On the first day of observation, there were 372 eggs in the 56 leaf pairs 

tagged. On the second day, the number ofeggs increased to 619. On the third day, 

a total of 750 eggs and 93 first instar larvae were found. There were no further 

additions on subsequent days. It can be seen that the maximum number of eggs 

was laid on the first day. On the second and third days, the numbers of eggs laid 

were almost equal- 269 and 268 respectively. Observations on 56 leaf pairs 

concluded that egg-laying can happen on the same leaf for at least three 

consecutive days. The population ofmoths that laid eggs on all the three days can 

either be the same group of moths arriving each day from a place of aggregation 

or different groups ofmoths. 


Cage observations confirmed the earlier known information on pre-mating 

period. The characteristic calling behaviour of the moths could be observed in the 

cage, which could notbe seen in the earlier laboratory studies. Observations made 

outside the cage gave new information on one of the most mortality prone life 

stage of the insect- the time immediately following adult emergence from the 

pupa. Attraction of birds to the emergence site suggests that they are responding 

to some cue, possibly scent emanating during adult eclosion. 

75

Aggregations of the moths were predominantly found on hillocks at 

Thannikkadavu and Ambalakkunnu during the daytime. "Hilltoping" is a 

phenomenon exhibited by many species of lepidopterans to aggregate from a wide 

area (Brown and A1cock, 1990; Pinheiro, 1990). It has been argued that this 

behaviour facilitates the insects from a wide area to reach the top of one or other 

hillock. Aggregation of moths within the hilltop can be mediated through 

chemical means. Thus "hilltoping" can help the formation of a moth aggregation. 

The aggregations, which were observed, consisted ofmoths emerged on different 

days as indicated by the simultaneous presence offresh and old moths. 

Observations on the dispersal of adult defoliator moths indicated two 

distinct types of movement. The first was the unidirectional movement observed 

outside the plantation and the second was the dispersal type of movement found 

within the plantation. While the first type of movement was noticed during 

daytime, the second type ofmovement was found only during dawn and dusk. 

Even though only a small number of moths could be collected during 

flight, it indicates that during unidirectional flight, the group consists 

predominantly of mated females. It can be inferred that mating occurs before the 

mass movement of the moths. However, males are not absent from the migrating 

group. The aggregations that were detected were also predominantly consisting of 

females. There is high chance of mortality during mass movements. Therefore, 

mating before the movement reduces the risk of not finding enough number of 

males at the destination to mate with the surviving females. 

In this study, egg-laying has been found to occur for three continuous days 

in the same plantation. In the earlier studies (Mohanadas, 1995) it had been noted 

that at any given time during the larval period of the insect, anyone among the 

five instars will be dominating even though simultaneous presence of at least 

three instars were noticed. This observation had led to the conclusion that the 

76

insect lays eggs only on a single day and the presence of instars other than the 

dominating one was due to the difference in developmental time ofthe insect. The 

present study shows that egg-laying on different days can also lead to the 

presence of more than one instar. This oviposition behaviour need not necessarily 

be typical as the present observations were made during the month of July in a 

young plantation. More observations are needed on the oviposition behaviour of 

the immigrant moths during the early outbreak period. 

Based on the results of this study, the sequence of events from the 

emergence ofmoths to the start ofthe next outbreak can be described as follows: 

The newly emerged moths after a period of about one hour move by active 

flight to form an aggregation. Moths that have emerged from different stands 

and/or those emerged on different days form part of the same aggregation. After 

mating, the aggregation moves to a new habitat. Post migratory aggregation can 

also bring in moths from different stands or those emerged on different days. Both 

males and females will be part of this group but females are predominant. It 

seems that the aggregation remains until the completion ofegg-laying. 

The behaviour of the defoliator moths proposed above can explain many 

of the characteristics ofH. puera population dynamics. Incidence of outbreaks in 

different patches at the same time can be caused by short-range movement of a 

group of moths during the period of egg-laying. Similarly, the simultaneous 

occurrence of more than one stage of the insect at any particular stand can be 

explained by the fact that the insect lays eggs on the same stand more than once. 

It is not certain whether the eggs laid on a single stand on different days is by the 

same group ofmoths or by different groups. Long-range movement oflarge group 

of moths outside teak plantations observed in this study could cause sudden 

outbreaks in plantations. 

77

CHAPTER VII 

DEVELOPMENT OF OUTBREAK MONITORING METHODS 


Early detection of defoliator population is necessary to adopt timely 

measures to prevent damage. The eggs of the defoliator are quite small which 

can only be detected on careful observation. The first instar larvae, which 

emerge from the eggs, are also small and they feed by scraping the foliage and 

therefore neither the larva or the damage caused is visible from ground. The 

damage can be detected visually only when the second instar larvae cut the 

leaf and make folds out of it. By this time at least three days would have 

elapsed from the start of the outbreak. Since egg-laying is the first step in the 

initiation of an outbreak, the best indicator of a forthcoming infestation is the 

presence of moths. It is known that H .puera moths can be monitored by light 

traps (Vaishampayan and Bahadur, 1983). However, a major limitation in the 

use of light traps is the necessity to have electricity at the site of operation. 

This is often difficult, particularly in teak plantations. To surmount this 

difficulty, attempts have been made to use car batteries as the source of 

electricity. However, the need to recharge the batteries at frequent intervals, 

which is either cumbersome or not practicable under some situations, is a 

serious handicap. In addition, the light intensity of battery operated lamps will 

vary depending on the battery charge. Incandescent (tungsten filament) bulbs 

of 100 to 200 watts are generally used in light traps. The Pennsylvanian trap 

uses a fluorescent tube. An increase in the intensity of light usually results in 

increased trap catches. Use of ultraviolet light also increases the trap catches, 

particularly of moths (Southwood, 1976). "Black-light" tubes emitting 

portions of the ultraviolet spectra, not harmful to the human eye, have recently 

been developed and are now commercially available. Since they are much 

more attractive than incandescent lamps, tubes of lower wattage can be used. 

In this study, a solar-powered light trap was developed for operating in 

plantations and the correspondence of outbreaks and light trap catches was 

tested.

Since it was found that the correspondence between light trap catch 

and outbreaks was not adequate, attempts were made to develop an outbreak 

monitoring method by combining light trap and visual observations. 

7.2 METHODS 

The light trap was developed using a black-light tube, powered by a 

battery. The battery is charged during the daytime, using sunlight, through a 

photovoltaic system. The light trap system consisted of three sub-units: (I) the 

trap, (2) the collection cage, and (3) the Solar photovoltaic (SPY) system. 

(1) The trap 

Details ofthe trap system are given below (see Fig.7.1). 

The basic trap unit is similar in design to the Pennsylvanian trap. It consists 

of a framework made of two flat iron rings, 30 cm in diameter, with cross 

arms in the middle, and connected together with 4 iron rods. A tube holder is 

fixed at the centre of the cross arms of each ring, to hold a fluorescent tube. 

Aluminium channels fixed in the cross arms serve to hold 4 baffles, 60 cm tall 

and 12 cm wide, made of 4 mm thick transparent perspex. The baffles can be 

removed to replace the tube. A funnel, 30 cm diameter at top, made of 20 

gauge smooth aluminium sheet, is fixed to the bottom ring of the framework. 

The tail of the funnel extends into a collection cage. Alternatively, the tail of 

the funnel passes through a hole cut in the centre of a stainless steel bottle cap 

to which a collection bottle can be screwed. The top ring supports a conical 

aluminium hood 45 cm in diameter at bottom, to protect the trap from rain. 

Insects are attracted by light emitted by a 20 W black-light fluorescent tube, 

60 cm long. It works on single phase AC electric supply of 230 Y, 50 Hz and 

emits light rays of low frequency (Wavelength 300 to 400 nm), not harmful to 

human beings. 

(2) The Collection Cage 

The Collection cage is a walk-in-cage, 180 x 180 x 210cm, made of angle 

iron frame and covered with nylon netting fixed with nuts and bolts. The cage 

is placed on a basement plastered with cement. 

79

-.-.-._ --- _--. Black light tube 

--._ -._ _ _. Baffle 

......__.-•._-._.-.--- _- Tube holder 

- " --.---- ,,- Collectionfunnel 

..."..__ __._._. Collection tube 

- _.__.._.__._- Walk-in cage 

"....__._..._._....- Battery box 

Fig.7.1. Diagram showing the prototype of the light-trap. 

(3) The Solar Photo-voltaic System 

The Solar photo-voltaic system (SPV) system was developed with the 

help of ANERT (Agency for Non-conventional Energy and Rural 

Technology), Trivandrum. It consists of (i) 4 numbers of 30W (nominal) SPY 

panels, (ii) a l2V 60 AH storage battery, and (iii) an electronic control unit. 

The electronic control unit has two sections, namely charge controller and 

inverter. The charge controller ensures that the battery is neither over-charged 

nor over-discharged, with cut-ofT voltages at B .8V and 1O.5V respectively. 

The inverter operates at 20 kHz with a secondary voltage of 200V on load. 

The Solar panels are mounted on a steel pole, 2.6- ID in height and 7.5- cm. in 

diameter. The battery and electronic control unit are housed in a box mounted 

on the pole to protect them from rain and dust. 

The light-trap was established on top of a hillock at Kariem 

Muriem. This hillock is located in the centre of the teak plantations in this 

80

area. The trap was operated for a period of 16 months from 23 June 1993 to 15 

October 1994. Daily collections of defoliator moths were made during the 

period 1993-94. The outbreaks, which occurred during this period, were 

mapped and the dates of start of these outbreaks were estimated as described 

in Chapter 3. During this period, all the outbreaks that occurred at Kariem 

Muriem were mapped and based on larval samples collected, the starting date 

of each infestation was determined as described in Chapter 3. It was examined 

whether moths were collected in the trap during the start of an outbreak so as 

to judge if it could be used as a monitoring device. 

7.3. RESULTS 

The solar light trap developed was found satisfactory. Based on the 

prototype model described above, a more convenient portable model was 

designed with the following modifications: 

1. The number ofsolar panels was reduced from 4 to 1. 

2. The steel pole was reduced in size and was made collapsible. 

3. The walk-in cage was replaced with a smaller cage. 

The trap catch during the 16-month period is given in Appendix Band 

depicted in Fig.7.2. along with the time of occurrence of outbreaks at Kariem 

Muriem. Distinct period of abundance of moths in the field can be identified. 

The period immediately after the establishment of the trap (June, 1993) was a 

time of high abundance of moths. There was a decline in trap-catch during 

August. There were further increases in the number of moths during the first 

weeks of September and October. Only very few moths were collected during 

November (2 moths) and December (1 moth). 

In 1994, no moths were collected during January, February and March. 

The first moth was collected on 2 April. Large number of moths were trapped 

during the moths of May, June and July beyond which no moths were 

collected until the end of the study in October. The correspondence between 

the trap catch and the initiation ofan outbreak is depicted in Table 7.1. 

81

Table 7.1. Correspondence between light-trap catch and incidence ofoutbreak: 

Sl.No. of Starting Cwnulative Distance Area Whether 

outbreak date of number of between infested locally 

which outbreak moths trapped light-trap (ha) emerged 

occurred (date ofegg during a 5 day and the moths are 

during the laying) period prior infested present 

period of to start of site (km) 

light-trap outbreak 

operation 

1 7 Julv 1993 145 0.05 52.0 Yes 

2 27 August 0 0.5 36.1 No 

1993 

3 I 74 2.0 35.7 No 

September 

1993 

4 4 April 1 0.5 12.2 No 

1994 

5 12 May 9 0.05 472.5 No 

1994 

6 3 June 1994 596 3.0 75.0 Yes 

7 12 June 0 0.05 435.3 No 

1994 

The first outbreak: occurred on 7 July, 1993 at an area of 52 ha. which 

was 0.05 km away from the light-trap. A total of 145 moths were collected 

during the five day period prior to 7 July. There were locally emerged moths 

during this period at Kariem-Muriem from an outbreak which occurred on 10 

June. The second outbreak: was on 27 August 1993 covering an area of36.1 ha 

at a distance of0.5 km from the light-trap. No moths were collected during the 

days prior to this date. There were no locally emerged moths at Kariem 

Muriem during this period. The third outbreak: also occurred when there were 

no locally emerged moths but 74 moths were collected during the days prior to 

the start of the outbreak. This outbreak occurred at a distance of 2 km away 

from the light-trap and covered 35.7 ha. The fourth outbreak occurred on 4 

April 1994 at distance of 0.5 km from the light-trap and covered an area of 

12.2 ha. Only one moth was collected and there were no locally emerged 

moths. A major outbreak occurred on 12 May 1994 covering an area of472.5 

ha and 0.05 km away from the light-trap. Only 9 moths were collected prior to 

this outbreak. No locally emerged moths were present during this period. The 

next outbreak extended to an area of 75 ha at a distance of 3 km from the 

light-trap. A total of 596 moths were collected on days preceding the start of 

83

light-trap. A total of 596 moths were collected on days preceding the start of 

this outbreak. Locally emerged moths were present during this period. No 

moths were collected prior to the last outbreak which was 0.05 km away from 

the trap and which extended to 435.3 ha. There were no locally emerged 

moths during the period. 

It can be seen that out of seven outbreaks which occurred during the 

study period, occurrence of two outbreaks (s1.nos. 2 and 7) could not be 

detected by light-trap catch eventhough the sites infested were very close (0.5 

and 0.05 km respectively) to the trap. Very few moths were collected during 

the 4 th and 5 th outbreaks (l and 9 respectively) which also occurred near to the 

light-trap (0.5 and 0.05 km respectively). Only the occurrence of three 

outbreaks (s1.nos. 1,3 and 6) among the total seven was well indicated by trap 

catch. Majority of moths were collected during the l" and 6 th outbreaks (145 

and 596 respectively) during which moths were emerging locally from earlier 

outbreaks. Only the occurrence of the third outbreak was well indicated (74 

moths were collected) by trap-catch when no locally emerged moths were 

present. Thus when no locally emerged moths were present, only one of the 

five impending outbreaks was predictable by a high trap catch. 


This study indicates that eventhough the moths were collected in the 

light-trap during the period of the year when teak defoliator outbreaks are 

prevalent, it does not always collect moths, which arrive in the plantation for 

egg laying. Large number of moths was collected while they were emerging 

from the plantations nearby. But the trap was unable to attract and collect 

moths every time they arrived at the plantation for egg laying. The reason is 

not well understood but it appears that the moth aggregation that arrives for 

egg laying respond primarily to the chemical signals from the host plant. 

This study has shown that the light-trap cannot be relied upon as an 

outbreak detection device. Detection of an outbreak needs ground 

observations. These observations can be limited to areas, which have tender 

84

foliage. If folds are detected, observations have to be made' in transects 

radiating in four directions from the site where folds are detected. At these 

transects, tender leaves have to be closely observed for the presence of 

defoliator eggs or first instar larvae. Eggs are oval and white. The first instar 

larvae will be feeding by scraping the leaves near the veins. This observation 

is needed because egg laying can occur for more than one day and sites with 

eggs and first instar larvae can go unnoticed in an observation from ground. 

85


CHAPTER VIII 

POPULATION DYNAMICS OFH. PUERA - 

A SYNTHESIS OFAVAILABLE INFORMAnON 

The study of population dynamics is an old discipline that antedates 

the modern science of ecology (Cappuccino, 1995). Insects have been a much 

researched group owing to their short-life span and role as pests. Of the 

various pests, those that dwell in forests have attracted much interest since 

they occupy relatively natural environment as compared to those in many 

agricultural systems (Berryman, 1986). H. puera outbreaks that occur in teak 

stands with a normal rotation period of60 years present a unique case to study 

population dynamics. Moreover, teak defoliator outbreaks occur more than 

once every year compared to the 8-11 years frequency seen in the case of 

many temperate species of forest defoliators (Myers, 1998). 

According to a classification scheme based on the spread of outbreaks 

(Berryman, 1987), teak defoliator outbreaks have been recognized as 

belonging to the eruptive type (Nair, et al., 1994). The main characteristics of 

insects that display eruptive outbreaks is that their populations remain at 

relatively stable levels for long periods but then suddenly erupt to very high 

densities. These eruptions usually begin in particular localities (epicentres), 

and then spread over large areas. This theory implied that controlling the 

initial epicentres could lead to suppression oflarge-scale outbreaks. 

The epicentre hypothesis has practical value, if it is proved that 

progenies of epicentre populations cause the large-scale outbreaks. This needs 

simultaneous observation in large areas and precise information on the time of 

start of each outbreak. Then, based on the generation time needed for each 

population, we can determine whether the large-scale outbreaks could 

originate from initial epicentres. Such an attempt was made in the present 

study. Flight characteristics of moth were also studied to explain the 

population dynamics exhibited by the insect. Since the spatial scale of the

study has considerable bearing on the perception ofthe phenomena (Solbreck, 

1995), observations were carried out in a large geographical area. 

This chapter attempts to synthesize the earlier available information 

and those generated in the present study in the light of recent advancement in 

theory on insect population dynamics. Important aspects influencing the 

population dynamics of the insect, namely aggregation, flight, origin of 

outbreaks, and the pattern of spread of outbreaks are discussed and an attempt 

is made to develop a theoretical framework appropriate for describing teak 

defoliator outbreaks. 

8.2. AGGREGATION AND FLIGHT OF MOTHS 

It has been recognized that the tendency to aggregate is a characteristic 

ofoutbreak species of insects (Cappuccino, 1995). In the case ofH. puera, the 

sudden appearance of heavy infestation with thousands of larvae per tree, 

following a period of near absence of infestation had indicated that 

aggregation of moths occur prior to egg-laying at the site. Moth aggregations 

have been observed earlier within teak plantation and nearby natural forests 

(Nair, 1988). In the present study aggregations of moths were observed in teak 

plantations immediately before the plantation was infested (Chapter 6). These 

aggregations consisted of uneven aged moths of both the sexes. This indicates 

that an aggregation is composed of moths that emerged from different sites or 

moths that emerged from the same site on different days. The fact that 

aggregations were predominantly found on hillocks suggests that moths use 

topography as a guiding cue to form aggregation. 

Circumstantial evidence for short-range (Nair and 

Sudheendrakumar,1986) and long-range (Vaishampayan et al., 1987) 

movement of moths was obtained earlier based on independent studies at 

Kerala and Madhya Pradesh. In the present study, direct observations revealed 

two types offlight behaviour in H. puera. 

87

I. Dispersal flight: observed during dawn and dusk in all directions within 

teak plantations at the canopy level, 

2. Directional flight: not restricted to dawn and dusk. moths moving in the 

same direction in a swarm. 

8.3. ORIGIN OF OUTBREAKS 

This study showed that in 1993, within the nearly 9,000 ha teak 

plantations at Nilambur, the first outbreaks occurred during the month of 

February in a few small scattered patches. These patches varied in size from 

1.8 to 12 ha. These initial patches could originate in two possible ways: 

I. A change in behavior of endemic population of the insects within the area 

leading to aggregation and mass egg-laying. 

2. An influx ofmoths from a distant area 

Correlation between the time of occurrence of first outbreaks and pre 

monsoon showers has been observed earlier. Since there is no report on 

diapause in H. puera, rain cannot be a factor that triggers the emergence of 

moths. Even though new flushes come up profusely after the pre-monsoon 

showers, it is observed that tender foliage sufficient to support epidemic 

populations of the insect are present even before the pre-monsoon showers. It 

could be thought that the first rains could have an impact on the behaviour of 

moths, inducing them to aggregate. Alternatively, the wind system associated 

with the pre-monsoon showers can assist in long distance immigration of 

moths. The present data are not sufficient to prove anyone ofthe above. 

8.4 SPATIAL SPREAD OF OUTBREAKS 

The spread of outbreaks during a year within nearly 9,000 ha teak 

plantations at Nilambur can be summarized as follows: 

It was observed earlier that initial outbreaks occurred in small patches 

covering 0.5 - 1.5 ha in area which are widely separated. The present study 

showed that the initial epicentres could be much larger extending to a 

88

maximum of 13 ha (see Section 5.2, Chapter 5). It was noticed that these 

epicentres originate during the month of February. As discussed above, origin 

of epicentres remains an unresolved problem. During the months March and 

April, infestations occur in patches of a wide range of sizes (0.1 to 934 ha), 

which are still widely separated in space. Most of these outbreaks occur 

simultaneous with the emergence of moths from the populations during the 

first phase (the epicenter phase), but some populations occur when there are 

no locally emerged moths. Widespread outbreaks occur in a large number of 

patches during May and June. Progenies of populations, which occurred 

during the build-up phase, could cause all the outbreaks during this phase. 

While the life stage of the insect is uniform within an outbreak patch, there is 

considerable difference between patches. This could be because ofthe fact that 

moths emerged from different outbreaks that occurred during build-up phase 

cause these wide spread outbreaks. During July there is a reduction in both the 

number ofpatches infested and the size ofoutbreak patches. All outbreaks that 

occur during July could be explained as caused by progenies from earlier 

populations. Outbreaks during this period do not cause further outbreaks in the 

area even if tender foliage is present. This may be because of collapse of 

population due to natural mortality factors or the emigration ofmoths from the 

area. A few outbreaks covering around 1-40 ha occur during the period August 

- September. With respect to origin of these outbreaks, this phase resembles 

the period of epicentres. These outbreaks seldom cause subsequent outbreaks 

in the area. 

8.5. THE BACKGROUND TO WORKING TOWARDS THEORY 

The following specific details hitherto generated are relevant to 

explaining the observed dynamics ofteak defoliator outbreaks: 

1. At the global scale (encompassing all places were H. puera is present i.e., 

India, Myanmar, Sri Lanka, Indonesia, Papua-New Guinea, the Solomon 

islands, etc) the insect occurs in outbreak density at different places at 

different times. 

89

2. At a still lower regional scale (for example the Indian Subcontinent) there is 

a directional progression of outbreaks, which seems to be linked with the 

movement ofmonsoon wind system. 

3. At the local level (like the teak plantations of Nilambur), outbreaks occur 

only during some part of the year. During the outbreak period, infestations 

originate in a few small epicentres. Later, outbreaks spread td larger and 

larger areas. While most of the outbreaks could be caused by previous 

outbreaks, a few are not so. During the final phase of the outbreak period, a 

few outbreaks occur that can only be explained either as caused by long 

range migration of moths or by aggregation of moths from the endemic 

population. After this, the population density remains low until the next year 

when the sequence of outbreaks is repeated. Thus, the teak defoliator 

displays population cycles at a frequency ofone year. 

Various explanatory hypotheses have been proposed for the dynamics 

of forest lepidoptera that exhibit population cycles (Myers, 1988). These 

include: 

a) Variation in insect quality: Chitty (1967) proposed that density- 

related selection on genetically controlled variation in behaviour and 

physiology of animals could provide the basis for self regulation of 

populations. Experimental proof is still non-existent for this hypothesis 

(Myers, 1988). 

b) Climatic release hypothesis: Uvarov (1931) and Andrewartha 

and Birch (1954) proposed that weather and climate are major controlling 

factors of insect abundance. Climate can cause direct and indirect impact on 

population size, but the hypothesis remains untestable due to the difficulty in 

. differentiating the impact due to climate from that caused by other factors 

(Myers, 1998). 

90

c) Variation in plant quality I availability: Two hypotheses have 

been proposed based on the quality offood available for the insect. The first is 

that the quality of foliage may deteriorate following herbivore damage and 

thus act in a delayed density-dependent manner to reduce the population size. 

The other is that the nutritional quality of foliage improves following 

environmental stress from drought, waterlogging etc. so that the population 

size increases. The availability of food can also cause population cycles. In a 

deciduous tree like teak, it is probable that the absence of tender foliage during 

some part ofthe year can cause population decline ofthe herbivore. 

d) Disease susceptibility: This hypothesis proposes that as the 

population of an insect increases, individuals interact more frequently, thus 

allowing the transmission of disease. Stress associated with food limitation or 

poor weather could further accentuate the susceptibility to disease, which 

results in an epizootic. High mortality from disease selects for resistant 

individuals and the epizootic ends as the host density declines. Sub-lethal 

effects of disease may reduce vigor and fecundity for several subsequent 

generations. 

e) Metapopulation theory: Metapopulation is a set of local 

populations inhabiting spatially distinct habitat patches. A conceptually 

important assumption is that local populations have a significant risk of 

extinction (Moilanen and Hanski, 1998). The metapopulation is assumed to 

persist in a stochastic equilibrium between local extinctions and colonizations. 

Thus migration is a major driving force in metapopulation dynamics. 

When we attempt to explain the population dynamics of H. puera, it 

can be seen that no theory can explain it completely but all the theories 

provide insight into one or the other characteristics ofteak defoliator dynamics 

at the population level. The first theory regarding variation in insect quality is 

important since morphologically distinct larvae are found in the field which 

tempted the early workers to classify them as two distinct varieties (see 

Chapter II). The second hypothesis on climatic release is interesting in the 

91

Chapter Il). The second hypothesis on climatic release is interesting in the 

scenario where the teak defoliator outbreaks start immediately after the pre 

monsoon showers. The third hypothesis on variation In plant 

quality/availabitlity is also important since teak is a deciduous tree which 

sheds its leaves during winter, thereby creating a time when no food is 

available for the insect larvae. The fourth theory on disease susceptibility is 

also of interest owing to the recent discovery of the baculovirus, which can 

cause wide-spread epizootics and can persist within the host insect. Even 

though none of the above theories can be explicitly ruled out, the 

metapopulation theory in which space is identified as a determinant in 

regulating population dynamics appears to hold good in the case of Hipuera 

outbreaks. An attempt is made below to view the population dynamics ofteak 

defoliator in the light ofmetapopulation theory. 

8.6. AN EXPLANATORY MODEL FOR THE POPULAnON DYNAMICS 

OF H PUERA 

A schematic diagram showing the population dynamics ofH puera is 

given in Fig. 8.1. The figure is divided into two parts. The upper part above 

the dotted line indicates the metapopulation, which is the assemblage of 

several local populations. This metapopulation can extend to the total 

distribution area and may exist in a high density (epidemic) level in one or the 

other areas while it remains at low density (endemic) level at the other places. 

Even when the size of the metapopulation remains stable, at the local level, 

population density can shift between endemic and epidemic levels. Below the 

dotted line, the local population dynamics at a place like Nilambur is 

represented. 

At the local level, low density, endemic populations have been 

observed during the non-outbreak period. The insects are rare and distributed 

in a disperse manner throughout the plantation. Incidence of first outbreaks 

(epicentres) at a particular place (of several sq.krn in area) can occur either by 

the aggregation of local endemic population or by immigration of moths from 

92

away places. These populations seldom cause further outbreaks in the area and 

the local population declines to endemic level. This may be due to the 

reduction in food source since most of the leaves would be mature and not 

acceptable to the early larval instars. 

The graph in Fig 8.1. shows the temporal sequence of shifts in 

population density. At Nilambur the population density remains at endemic 

level during January and February (A). During late February or March, the 

initial epicentres occur causing an increase in population density (B). From 

April to July, the population remains at epidemic level, causing large 

outbreaks at different places (C). During August, the population reverts to 

endemic level (Az). During September - October further small-scale outbreaks 

occur (Bj) followed by a period of endemic population (A3) which extends to 

February, next year. 

At the local level a specific period can be identified during which 

control operations are feasible. Control of the epicentres and any new 

populations occurring during the build-up phase can theoretically prevent 

large-scale outbreaks. This would be an economical way of preventing 

outbreaks as compared to an attempt to control wide-spread outbreaks. The 

model also indicates that it will be impossible to prevent all outbreaks by this 

method of control since the outbreaks which occur during the final phase are 

independent from the sequence of outbreaks which occur during the early part 

of the year. But this study has shown that controlling the initial outbreaks can 

prevent outbreaks in nearly 78% of the total area under outbreaks. The model 

also indicates that since recolonization can occur from far away habitats, 

control operations have to be repeated every year. 

94

REFERENCES 

Ahmed, M. (1987) Relati ve resistance of various clones of Tectona grandis to 

teak defoliator, Hyblaea puera Cram. (Lepidoptera, Hyblaeidae) in 

south India. Indian Forester. 113(4): 281-286. 

Anderson, H., Durston, B. H. and Poole, M. (1970) Thesis and assignment 

writing. Wiley Eastern Limited, New Delhi: 135 pp. 

Andrewartha, RG. and Birch, 1. C. (1954) The distribution and abundance of 

animals, xv+782 pp. Chicago. 

ARC/INFO (1995) Understanding GIS- The ARCIINFO method: version 7 for 

UNIX and open VMS. 3 rd edition. John Wiley & Sons Inc., New York, 

NY-I0158. 723 pp. 

Basu-Chowdhury, J. C. (1971) Interim report on Konny Aerial Spraying 

Project. Appendix 3 In: Vasudevan K.G., Working plan for Nilambur 

Forest Division, 1967-68 to 1976-77. Govt.ofKerala, pp.165-168. 

Beeson, C. F. C. (1928) The defoliation of teak. Indian Forester. 54(4): 204 

215. 

Beeson, C. F. C. (1934) The biological control of teak defoliators. Indian 

forester. 60( 10):672-683. 

Beeson, C. F. C. (1941) The ecology and control of the forest insects ofIndia 

and neighbouring countries. Reprint 1961. Govt. of India publication, 

767 pp. 

Berryman, A. A. (1986) Forest insects: Principles and practice of population 

management. Plenum Press, New York. 279 pp.

Berryman, A. A. (1987) The theory and classification of insect outbreaks. In: 

Insect outbreaks (Eds. Barbosa, P. and Schultz, lC.). Academic Press, 

California, USA. 577 PP. 

Bourdillon, T. F. (1898) Insects attacking teak in southern India. Indian 

Forester. 24:126-129. 

Brown, W. D. and Alcock, 1 (1990) Hilltopping by the red admiral butterfly: 

mate searching along congeners. Journal of research on the 

Lepidoptera. 29 (1-2): 1-10. 

Capuccino, N. (1995) Novel approaches in the study of population dynamics. 

In: Population dynamics: new approaches and synthesis. Pp 3-16 (Eds. 

Cappuccino, Naomi and Price, Peter.W.) Academic Press Inc. London. 

Chitty, D. (1960) Population processes in the vole and their relevance to 

general theory. Canadian Journal ofZoology. 38: 99-113. 

Hole, R. S. (1904) Note on Hyb/aea puera. Indian Forester. 30( 1): 1-10. 

Kadambi, K. (1951) Teak defoliation. Indian Forester. 77(1):64-70. 

Khan, A. H. and Chatterjee, P. N. (1944) Undergrowth in teak plantations as a 

factor in reducing defoliation. Indian Forester. 70: 365-369. 

Liebhold, A. M. and Elkinton, J. S. (1989) Characterizing spatial patterns of 

gypsy moth regional defoliation. Forest Science. 35(2) pp 557-568. 

Liebhold, A. M., Rossi, R. S. and Kemp, W. P. (1993) Geostatistics and 

Geographic information systems in applied insect ecology. Annual 

review ofEntomology. 1993. 38: pp 303-27. 

Mackenzie, 1. M. D. (1921) Some notes on forest insect pests in Burma. Indian 

Forester. 47:309-317. 

96

Mathew, G. (1991) Biology, seasonal population trends and impact of the teak 

carpenter-worm Aleterogystia cadambae (Moore) (Lepidoptera: 

Cossidae). Annals ofEntomology. 9(2): 39-46. 

Mohanadas, K. (1995) Population trend of Hyblaea puera Cramer 

(Lepidoptera: Hyblaeidae) in teak plantations and the factors 

influencing it. Ph.D. thesis. Cochin University of Science and 

Technology, 159 pp. 

Moilanen, A. and Hanski, I. (1998) Metapopulation dynamics: Effects of 

habitat quality and landscape structure. Ecology. 79 (7): 2503-2515. 

Myers, 1. H. (1988) Can a general hypothesis explain population cycles of 

forest lepidoptera? In Advances in ecological research.Vol. 18, 

Academic Press Inc. (London) pp, 179-242. 

Nair, K. S. S. (1980) The problem of insect defoliation in teak- to spray or not 

to spray. In: Proceedings of second Forestry conference, Dehra Dun, 

India. 

Nair K. S. S. and Sudheendrakumar, V.V. (1986) The teak defoliator, Hyblaea 

puera: Defoliation dynamics and evidences ofshort-range migration of 

moths. In: Proceedings of Indian Academy of Sciences (Animal 

Sciences) 95 (1): 7-21. 

Nair, K. S. S. (1987) Life-history, ecology and pest status of the sapling borer 

Sahyadrassus malabaricus (Lepidoptera: Hepalidae). Entomon 12(2): 

167-173. 

Nair, K. S. S. (1988) The teak defoliator in Kerala, India. In: Forest Insects: 

Principles and practice of population management. (Ed. Berryman, 

A.A.) pp 267-289,Plenum Press, New York. 

97

Nair, K. S. S., Mohanadas, K and Sudheendrakumar, V. V. (1995) Biological 

control of the teak defoliator, Hyblaea puera (Lepidoptera: Hyblaidae) 

using insect parasitoids- problems and prospects. In: Biological control 

of social forest and plantation crop insects (Ed. Ananthakrishnan, T. 

N.). Oxford and !BR Publishing Co. Pvt. Ltd.. New Delhi, pp 75-95. 

Nair, K. S. S. and Mohanadas, K. (1996) Early events in the outbreak of the 

teak caterpillar, Hyblaea puera. International Journal of Ecology and 

Environmental Sciences 22: 271-279. 

Nair, K. S. S., Biju Babjan, Sajeev, T.V., Sudheendrakumar, V. V., Ali, 

Mohammed, M. I., Varma, R. V. and Mohanadas, K. (1996) Field 

efficacy of nuclear polyhedraosis virus for protection of teak against 

the defoliator, Hyblaea puera Cramer (Lepidoptera: Hyblaeidae). 

Journal ofBiological Control. 10: 79 - 85. 

Naif, K. SS., Sudheendrakumar, V. V., Varma, R. V., Chacko, K. C. and 

Jayaraman, K. (1996) Effect of defoliation by Hyblaea puera and 

Eutectona machaeralis (Lepidoptera) on volume increment of teak. In: 

Impact of diseases and insect pests in tropical forests (Eds. Naif, K. S. 

S., Sharma, J. K. and Varma, R. V.) pp. 257-273 KFRI, FORSPA. 

Nair, K. S. S., Sudheendrakumar, V. V., Mohanadas, K and Varma, R. V. 

(1997) Control of the Teak Defoliator- Past attempts and new promise. 

In: Teak (Eds. Chand Basha, S, Mohanan, C. and Sankar, S.) Kerala 

Forest Department and Kerala Forest Research Institute. p 274: 81-83. 

Naif, K. S. S., Sudheendrakumar, V. V., Mohanadas, K., Varma, R. V., 

Mathew, G., Koshy, M. P., and Kedhamath, S. (1997) Search for teak 

trees resistant to the defoliator, Hyblaea puera Cramer (Lepidoptera, 

Hyblaeidae) In: Ecology and evolution of plant-feeding insects in 

natural and man-made environments (Ed. Raman, A.) 109-122, 

International Scientific Publications, New Delhi. 

98

Sudheendrakumar, V. V. (1994) Reproductive behaviour of H.puera and the 

role of pheromones. In: Development of a management strategy for the 

100 

teak defoliator, Hyblaea puera. (Eds. Nair et al.), KFRI research report 

95. 121 pp. 

Uvarov, B.P. (1961) Quality and quantity in insect populations. Proc. Royal 

Entomological Society of Canada (C) 25: 52-59. 

Vaishampayan, S. M. and Bahadur, A. (1983) Seasonal activity of adults of 

teak defoliator Hyblaea puera and teak skeletonizer Pyrausta 

machaeralis (Pyralidae: Lepidoptera) monitored by light trap catch. In: 

Insect interrelations in forest and agro ecosystems (Eds. Sen-Sarma, P. 

K, Kulshrestha, S. K., Sangal, S. K.) Jugal Kishore & Co. Dehra Dun 

137- 146. 

Vaishampayan, S. M., Verma, R. and Bhowmick, A. K. (1987) Possible 

migration of teak defoliator, Hyblaea puera Cramer (Lepidoptera, 

Hybaeidae) in relation to the movement of the south-west monsoon as 

indicated by light trap catches. Indian Journal of Agricultural Science 

57(1): 41-46. 

Zacharias, V. J. and Mohandas, K. (1990) Bird predators of the teak defoliator 

Hyblaea puera. Indian Journal ofForestry 13(2):122-127.

APPENDIX A 

ALGORITHM: FOR COMPUTATION OF AUTOCORRELAnON INDICES 

The general notation used in spatial autocorrelation formulas and their interpretation 

is as follows: 

n - the total number ofcells in a layer: N rows x N columns 

ij - any two adjacent cells 

Zi - the value ofattribute ofthe cell i 

cij - the similarity of i's and j's attributes: (zi-zi 

wij - the similarity of i's and j's locations, Wij = 1 if cells i and j are directly adjacent 

and 0 other wise 

2 - the sample variance: (z, - zml / (n - 1) where Zm is the mean cell value for 

o the grid 

In the terms ofthe above notation, spatial autocorrelation is a measure ofthe attribute 

similarities in the set of cij with the locational similarities of the set of Wij. and then 

summing the results into a single index. 

The formula for calculating the Geary index is: 

c = W C ij I( 2 (w )( 

U ij 

where. 

w=4*n 

ij 

The formula for calculating the Moran index is 

where, 

LLWij:::::4*n 

z - Z m ) 2 ) I( n - 1)) 

i 

\01

Date *No. of Incidence Local Date *No. of Incidence local 

moths of emergence moths of emergence 

rapped outbreak of moths trapped outbreak of moths 

7.9.93 5 Nil Nil 26.10.93 0 Nil Nil 

8.9.93 7 Nil Nil 27.10.93 0 Nil Nil 

9.9.93 4 Nil Nil 28.10.93 0 Nil Nil 

10.9.93 2 Nil Nil 29.10.93 0 Nil Nil 

11.9.93 0 Nil Nil 30.10.93 0 Nil Nil 

12.9.93 N.D. Nil Nil 31.10.93 1 Nil Nil 

13.9.93 0 Nil Nil 1.11.93 0 Nil Nil 

14.9.93 0 Nil Nil 2.11.93 0 Nil Nil 

15.9.93 0 Nil Nil 3.11.93 0 Nil Nil 

16.9.93 0 Nil Nil 4.11.93 0 Nil Nil 

17.9.93 0 Nil Nil 5.11.93 N.O. Nil Nil 

18.9.93 0 Nil Nil 6.11.93 N.D. Nil Nil 

19.9.93 N.D. Nil Yes 7.11.93 0 Nil Nil 

20.9.93 0 Nil Yes 8.11.93 0 Nil Nil 

21.9.93 0 Nil Yes 9.11.93 0 Nil Nil 

22.9.93 N.D. Nil Yes 10.11.93 0 Nil Nil 

23.9.93 0 Nil Yes 11.11.93 0 Nil Nil 

24.9.93 0 Nil Yes 12.11.93 1 Nil Nil 

25.9.93 0 Nil Yes 13.11.93 0 Nil Nil 

26.9.93 0 Nil Yes 14.11.93 N.D. Nil Nil 

27.9.93 1 Nil Yes 15.11.93 0 Nil Nil 

28.9.93 1 Nil Yes 16.11.93 0 Nil Nil 

29.9.93 2 Nil Yes 17.11.93 0 Nil Nil 

30.9.93 0 Nil Yes 18.11.93 1 Nil Nil 

1.10.93 0 Nil Yes 19.11.93 0 Nil Nil 

2.10.93 0 Nil Yes 20.11.93 N.O. Nil Nil 

3.10.93 N.O. Nil Nil 21.11.93 0 Nil Nil 

4.10.93 0 Nil Nil 22.11.93 0 Nil Nil 

5.10.93 2 Nil Nil 23.11.93 0 Nil Nil 

6.10.93 7 Nil Nil 24.11.93 0 Nil Nil 

7.10.93 3 Nil Nil 25.11.93 0 Nil Nil 

8.10.93 2 Nil Nil 26.11.93 0 Nil Nil 

9.10.93 2 Nil Nil 27.11.93 0 Nil Nil 

10.10.93 22 Nil Nil 28.11.93 N.D. Nil Nil 

11.10.93 4 Nil Nil 29.11.93 0 Nil Nil 

12.10.93 2 Nil Nil 30.11.93 0 Nil Nil 

13.10.93 0 Nil Nil 1.12.93 0 Nil Nil 

14.10.93 2 Nil Nil 2.12.93 0 Nil Nil 

15.10.93 2 Nil Nil 3.12.93 0 Nil Nil 

16.10.93 0 Nil Nil 4.12.93 0 Nil Nil 

17.10.93 N.O. Nil Nil 5.12.93 N.D. Nil Nil 

18.10.93 0 Nil Nil 6.12.93 0 Nil Nil 

19.10.93 0 Nil Nil 7.12.93 1 Nil Nil 

20.10.93 0 Nil Nil 8.12.93 0 Nil Nil 

21.10.93 0 Nil Nil 9.12.93 0 Nil Nil 

22.10.93 0 Nil Nil 10.12.93 0 Nil Nil 

23.10.93 0 Nil Nil 11.12.93 0 Nil Nil 

24.10.93 N.D. Nil Nil 12.12.93 N.D. Nil Nil 

25.10.93 0 Nil Nil 13.12.93 0 Nil Nil 

* N.D. indicatesdays on which the light-trapwas not operated. 

103

Date *No. of Incidence Local Date *No. of Incidence Local 


trapped outbreak ofmoths trapped outbreak ofmoths 

14.12.93 0 Nil Nil 1.2.94 0 Nil Nil 

15.12.93 0 Nil Nil 2.2.94 0 Nil Nil 

16.12.93 0 Nil Nil 3.2.94 0 Nil Nil 

17.12.93 0 Nil Nil 4.2.94 0 Nil Nil 



20.12.93 0 Nil Nil 7.2.94 0 Nil Nil 

21.12.93 0 Nil Nil 8.2.94 0 Nil Nil 

22.12.93 0 Nil Nil 9.2.94 0 Nil Nil 

23.12.93 0 Nil Nil 10.2.94 0 Nil Nil 

24.12.93 0 Nil Nil 11.2.94 0 Nil Nil 




28.12.93 0 Nil Nil 15.2.94 0 Nil Nil 

29.12.93 0 Nil Nil 16.2.94 0 Nil Nil 

30.12.93 0 Nil Nil 17.2.94 0 Nil Nil 

31.12.93 0 Nil Nil 18.2.94 0 Nil Nil 


2.1.94 N.O. Nil Nil 20.2.94 N.O. Nil Nil 


4.1.94 0 Nil Nil 22.2.94 0 Nil Nil 

5.1.94 0 Nil Nil 23.2.94 0 Nil Nil 

6.1.94 0 Nil Nil 24.2.94 0 Nil Nil 

7.1.94 0 Nil Nil 25.2.94 0 Nil Nil 

8.1.94 0 Nil Nil 26.2.94 0 Nil Nil 


10.1.94 0 Nil Nil 28.2.94 0 Nil Nil 

11.1.94 0 Nil Nil 1.3.94 0 Nil Nil 

12.1.94 0 Nil Nil 2.3.94 0 Nil Nil 

13.1.94 0 Nil Nil 3.3.94 0 Nil Nil 


15.1.94 0 Nil Nil 5.3.94 0 Nil Nil 


17.1.94 0 Nil Nil 7.3.94 0 Nil Nil 


19.1.94 0 Nil Nil 9.3.94 0 Nil Nil 

20.1.94 0 Nil Nil 10.3.94 0 Nil Nil 

21.1.94 0 Nil Nil 11.3.94 0 Nil Nil 

22.1.94 0 Nil Nil 12.3.94 0 Nil Nil 



25.1.94 0 Nil Nil 15.3.94 0 Nil Nil 

26.1.94 0 Nil Nil 16.3.94 0 Nil Nil 

27.1.94 0 Nil Nil 17.3.94 0 Nil Nil 

28.1.94 0 Nil Nil 18.3.94 0 Nil Nil 

29.1.94 0 Nil Nil 19.3.94 0 Nil Nil 



* N.O. indicates days on which the light-trap was not operated. 

104

Date "No. of Incidence Local Date "No. of Incidence Local 


trapped outbreak of moths trapped outbreak of moths 



24.3.94 0 Nil Nil 12.5.94 N.O. Yes Nil 

25.3.94 0 Nil Nil 13.5.94 4 Nil Nil 

26.3.94 0 Nil Nil 14.5.94 5 Nil Nil 


28.3.94 0 Nil Nil 16.5.94 0 Nil Nil 


30.3.94 0 Nil Nil 18.5.94 0 Nil Nil 

31.3.94 0 Nil Nil 19.5.94 0 Nil Nil 


2.4.94 1 Nil Nil 21.5.94 0 Nil Nil 


4.4.94 0 Yes Nil 23.5.94 0 Nil Nil 

5.4.94 0 Nil Nil 24.5.94 0 Nil Nil 

6.4.94 0 Nil Nil 25.5.94 0 Nil Nil 

7.4.94 1 Nil Nil 26.5.94 0 Nil Nil 

8.4.94 0 Nil Nil 27.5.94 2 Nil Nil 

9.4.94 0 Nil Nil 26.5.94 0 Nil Nil 


11.4.94 1 Nil Nil 30.5.94 116 Nil Nil 

12.4.94 0 Nil Nil 31.5.94 327 Nil Nil 

13.4.94 0 Nil Nil 1.6.94 37 Nil Yes 

14.4.94 N.O. Nil Nil 2.6.94 53 Nil Yes 

15.4.94 0 Nil Nil 3.6.94 63 Yes Yes 

16.4.94 3 Nil Nil 4.6.94 37 Nil Yes 

17.4.94 N.D. Nil Nil 5.6.94 N.O. Nil Yes 

18.4.94 0 Nil Nil 6.6.94 17 Nil Yes 

19.4.94 2 Nil Nil 7.6.94 O· Nil Nil 

20.4.94 1 Nil Nil 8.6.94 0 Nil Nil 

21.4.94 0 Nil Nil 9.6.94 0 Nil Nil 

22.4.94 0 Nil Nil 10.6.94 0 Nil Nil 

23.4.94 0 Nil Nil 11.6.94 0 Nil Nil 

24.4.94 N.O. Nil Nil 12.6.94 0 Yes Nil 

25.4.94 0 Nil Yes 13.6.94 0 Nil Nil 

26.4.94 0 Nil Yes 14.6.94 0 Nil Nil 

27.4.94 0 Nil Yes 15.6.94 0 Nil Nil 

28.4.94 0 Nil Yes 16.6.94 0 Nil Nil 

29.4.94 0 Nil Yes 17.6.94 0 Nil Nil 

30.4.94 0 Nil Yes 18.6.94 0 Nil Nil 

1.5.94 N.D. Nil Nil 19.6.94 N.O. Nil Nil 

2.5.94 0 Nil Nil 20.6.94 1 Nil Nil 

3.5.94 0 Nil Nil 21.6.94 0 Nil Nil 

4.5.94 0 Nil Nil 22.6.94 0 Nil Nil 

5.5.94 8 Nil Nil 23.6.94 4 Nil Yes 

6.5.94 8 Nil Nil 24.6.94 9 Nil Yes 

7.5.94 0 Nil Nil 25.6.94 1 Nil Yes 

6.5.94 N.O. Nil Nil 26.6.94 0 Nil Yes 

9.5.94 9 Nil Nil 27.6.94 1 Nil Yes 

.. N.O. indicates days on which the light-trap was not operated.

Date "No. of Incidence local Date "No. of Incidence Local 

moths of outbreak emergence moths of emergence 

trapped of moths trapped outbreak of moths 


5.10.94 0 Nil Nil 11.10.94 0 Nil Nil 


7.10.94 0 Nil Nil 13.10.94 0 Nil Nil 

8.10.94 0 Nil Nil 14.10.94 0 Nil Nil 

9.10.94 0 Nil Nil 15.10.94 0 Nil Nil 

• N.O. indicates dayson whichthe light-trap was not operated.

Spatial dynamics of teak defoliator (Hyblaea puera Cramer) - Cochin ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?