ECONOMIC THRESHOLD LEVEL AND MONITORING SYSTEMS: REALITY AND PRACTICALITY IN HELICOVERPA ARMIGERA (LEPIDOPTERA: NOCTUIDAE)

Indira Chaturvedi
Department of Zoology, Post Graduate College, Bilaspur, India

ABSTRACT

Field trials and sampling of commercial crops were carried out in Chhattisgarh region of India, to assess the impact of Helicoverpa armigera Hb. on chickpea (Cicer arietinum) at several larval densities and crop stages. The seasonal damage (i.e. pods damaged during crop growth) recorded was consistently higher than damage at harvest over almost the entire cropping season. The largest relative differences between both damage inspections occurred when infesting at early crop growth stages. Population estimates from destructive sampling and a rapid plant scouting method were related to pod damage, and a nominal economic threshold of one larva per plant was derived. In commercial demonstration trials, unsprayed controls and different insecticide treatments were used to obtain a range of H. armigera infestations and crop damage where standard calendar spraying practice was compared with no spraying, calendar-based applications maintained damage below 2.8%, when crops were sprayed on the basis of IPM scouting recommendations, the highest level of damage was 2.1% despite larval populations exceeding two per plant in five crops. Another study revealed that total and marketable yield in the trap-cropped plot was higher and damaged pods lower than in the monocropped chickpea. This article describes how preliminary studies of egg and larval distributions and damage assessment trials were used to develop an initial estimate of a threshold that would maintain damage below the 5% commercially acceptable level.

Key words: Helicoverpa armigera, Cicer arietinum, Economic threshold level.

INTRODUCTION

H. armigera (Lepidoptera: Noctuidae) is a wide spread polyphagous pest species of worldwide economic importance on many agricultural and horticultural crops [6,18]. This pest has been recorded feeding on 182 plant species across 47 families in the Indian subcontinent, of which 56 are heavily damaged and 126 are rarely affected [8]. Losses due solely to this pest of up to Rs.10, 000 million have been reported in crops like: cotton, pigeonpea, chickpea, groundnut, sorghum, pearl millet, tomato, and other crops of economic importance [11]. An extensive survey of agricultural fields revealed that extent of damage to crop and the consequent losses in yield due to this pest vary considerably from location to location and from season to season as agro-climatic conditions influence qualitative and quantitative composition of the pest complex. The distinct seasonal fluctuations in populations [16] have provided opportunities to restrict the application of insecticides to periods when they are necessary for control of the pest. Identifying these periods requires knowledge of economic thresholds and pest monitoring systems. In past decades, unreasoned or systematic (calendar spraying) chemical control has been progressively replaced by integrated pest management (IPM) programmes in India, although in the most cropping situations chemical control remains the chief tool to achieve successful control of H. armigera. Implementation of reliable economic-injury levels (EILs) and economic thresholds (ETs) for decision-making in IPM programmes leading to reduced insecticide sprayings is a crucial task, which will not only minimise chemical residues and environmental impact but also will doubtless improve the biological control exerted by native larval parasitoids [7,14].

Larvae may attack chickpea crops from transplanting until maturity but the most sensitive growing period coincides with the most attractive crop stages for ovipositing females, from early to last flowering. The predilection of this moth species for the harvestable fruiting parts, high polyphagy, wide geographical range, mobility, migratory potential, facultative diapause, high fecundity and propensity to develop resistance to insecticides are the many factors that contribute strongly to its pest status [2,5,6,15,16,18,19].

In this paper, a rapid plant scouting technique is described and used in conjunction with the nominal thresholds to define economic thresholds for commercial chickpea crops. This scouting technique and economic thresholds were evaluated as part of an experimental IPM program for processing chickpea crops.

MATERIAL AND METHODS

Fields trials and sampling of commercial crops were carried out in all chickpea growing areas of Chhattisgarh region in India during 2003-2005. Effects of crop growth stage, larval density (including the zero infestation level) and their interaction on either % seasonal damage, % damage at harvest and yield were determined for each year through Model I two-way ANOVAs following a randomised-complete-block design by using Systat [13] statistical software. Damage recorded in control plots (zero larvae per plant) were used to correct data within blocks at the remaining infestation levels by using the Abott [1] formula, when required.

Destructive sampling was used to obtained preliminary estimates of the relationship between egg and larval populations and pod damage in the field trials carried out during 2003-2005 as described in Cameron and Walker [4]. In these trials, unsprayed controls and two different insecticide treatments, viz endosulfan (94% w/w; Excel Industries, India) and monocrotophos (73% w/w; Khatau Junker Ltd, India) were used to obtain a range of H. armigera infestations and crop damage. Crops chosen for treatments were (1) Annigeri, (2) ICC 506 EB, (3) ICCV 37, (4) Vijay, (5) Venus, (6) ICCV 2, (7) Viswas, (8) ICCC 4, (9) Ram, (10) Punic, (11) Pant G 114 and (12) BG 372. The peak infestation by eggs and larvae (mean per plant) in the period of growth was related to the level of pod damage in the crops at harvest, using ANOVA to compute least significant differences among treatments based on the pooled error mean square [12]. Minor damage that had healed and would cause no loss of yield was omitted from the damage score. These results provided a nominal threshold that was used for the development of plant scouting in commercial crops. Observations were recorded on different growth and yield parameters. The data obtained from the experiments was analyzed statistically.

The 1-min scouting technique and economic thresholds were evaluated in field trials at the Research Station in 2004. This scouting technique was compared with destructive sampling using plant samples on two sampling occasions. Plants were scouted first and then destructively sampled. Twenty sample plants were randomly chosen in separate rows of each experimental block. The relationship of infestation levels to damage as determined from the experimental sampling, and the calibration of rapid scouting techniques were used to derive economic thresholds for rapid scouting of two eggs or one larva per plant. These thresholds were then used in subsequent experimental trials and commercial evaluations. The four crops, viz (1) Annigeri, (2) ICC 506 EB, (3) ICCV 37 and (4) Vijay were sown on 14 November and the four crops, viz (1) Venus, (2) ICCV 2, (3) Viswas and (4) ICCC 4 were sown on 14 December with three replications. Plant scouting using the 1-min technique for predicting pod damage was evaluated in unsprayed areas of commercial fields in four early and four late-planted crops in 2003-2004 and 2004-2005.

In 2004-2005 we performed a laboratory cage study and preliminary field study with the intention of developing cotton as a trap crop to combat H. armigera. The field experiment consisted of two large plots, one planted to only chickpea and the other planted to chickpea interspersed with rows of cotton. The crop was grown by standard cultural practices. No insecticide was used on the chickpea plants. From the start of chickpea flowering, we observed each chickpea plant from both blocks and cotton plants from the trap-crop block and recorded the number of H. armigera eggs laid. Three such observations were made at an interval of 15 days. At podding stage chickpea pod for H. armigera larval damage was observed five times at an interval of 2 weeks. The optimum allocation of sampling effort required to determine the mean number of larvae per plant in commercial fields was analyzed by comparing the variance and sampling costs between plants with those between quadrants, using data collected from commercial sites in 2003-2004 and 2004-2005.

RESULTS AND DISCUSSION

Seasonal damage and damage at harvest. Significant year-by-larval density and year-by-crop stage interactions were found in most cases, data from each cropping year were analysed separately using two-way ANOVAs. Results showed that both larval density and crop stage significantly affected seasonal damage, damage at harvest and yield, but the occurrence of significant larval density-by-crop interactions for all three variables indicated that the effect of larval density on pod damage and yield depended on the date of infestation relative to plant growth stages. Significant positive correlations between % seasonal damage and % damage at harvest were obtained all crop growth stages. The effect of larval density on damage at harvest was substantially different in 2003-2004. A significant block effect was observed in 2004 when seasonal damage and damage at harvest were analysed suggesting that environmental, operational or other non-random factors were involved.

Implementation of economic thresholds. The present study revealed that the time of initiation of H. armigera infestation in the field influences the crop yield to a great extent. Egg and larval populations were generally low in early-planted crops and did not exceed either the egg or larval thresholds. In late-planted crops, populations were higher, scouting detected more larvae than eggs, and the larval threshold of one larva per plant was commonly exceeded. The egg threshold of two per plant was not exceeded (fig. 1, fig. 2).

Data obtained from the destructive sampling of the field trials carried out in 2003 and 2004 provided a preliminary indication of the relationship between larval populations and pod damage. Peak larval populations during this period were maintained below one per plant only in the endosulfan treatments, which suffered low damage compared with the untreated control. These results suggested that maintaining populations below one larva per plant would restrict damage to ≈ 5% and ensure the crop was acceptable for harvest. This nominal economic threshold provided the basis for further field trials.

Scouting technique with destructive sampling and IPM scouting program. The comparison of the 1-min scouting technique with destructive sampling performed in the 2004 trial showed that significantly more eggs were found by destructive sampling (1.28 per plant) than 1-min scouting (0.56 per plant) and approximately twice as many larvae were found in destructive samples (1.12 per plant) compared with scouting (0.52 per plant). The efficiency of the 1-min scouting technique used in our study relied on the preference of Helicoverpa spp. for laying their eggs on the terminal half of branches on the exterior of the plant [17]. This preference was cofirmed for H. armigera by our comparison of 1-min scouting and destructive sampling. Estimates of the variance components, from all crops sampled in 2004-2005, for the standard 1-min sampling scheme of 10 plants sampled in each of four quadrants per field demonstrated that the ‘between quadrant’ variance component was small (0.00296 for larvae) compared with the residual variance (0.0399), using square root transformed data.

Comparison of damage in sprayed and unsprayed areas of crops sampled in 2003-2004 (tab. 1) demonstrated that although calendar-based insecticide applications maintained damage below 2.8%, the applications were often unnecessary. In five of the seven crops, five to seven insecticide applications reduced damage by <2% compared with unsprayed controls. The pod damage threshold of 5% was exceeded in the unsprayed areas of three crops and with sprays this damage was avoided by the use of five to seven applications. The potential for reducing insecticide applications was demonstrated when the IPM scouting program was implemented in 2003-2004 and 2004-2005 and applications were reduced to a maximum of one (tab. 2) compared with five to seven applications in calendar sprayed crops.

There were significant relationships between maximum larval populations and pod damage (using square-root transformations) from unsprayed areas of the 12 commercial crops sampled in 2003-2004. In only one of these crops (in 2003-2004) did larval populations below the economic threshold of one per plant produce damage exceeding 10%, and in one instance exceeding 7%. In the following two seasons (2003-2004 and 2004-2005), when crops were sprayed on the basis of IPM scouting recommendations, the highest level of damage was 2.10% despite larval populations exceeding two per plant in five crops (tab. 2). Our results show that one correctly timed insecticide application in the IPM program was sufficient to keep damage below the 5%commercially acceptable level.

Laboratory cage study and trap cropping systems. As in the cage experiment, H. armigera laid substantially more eggs on monocropped chickpea than on trap-cropped chickpea (tab. 3), except for the first observation when the number of eggs per plant was similar. Results of the H. armigera damage to chickpea pod observed five times during the season are summarized in (tab. 4). Within each treatment, pest damage varied from row to row. On average, however, chickpea planted with cotton as a trap crop suffered substantially less damage than chickpea planted as a sole crop. Total and marketable yield in the trap-cropped plot were higher and damaged fruit lower than in the monocropped chickpea (tab. 5).

For the initial IPM implementation trials, the larval economic threshold of one larva per plant was retained on the basis of the mean regression from commercial fields and the results of the threshold experiment. As expected, results clearly indicate that by using a 1% economic threshold, the 2% commercial standard damage will not be reached in most cropping situations, while yield losses obtained would range from 5% to 7%, which in most cases may be acceptable. In this sense, the 1% economic threshold is just based on expected pod quality at harvest and not on yield-related economic considerations. These economic-injury levels were consistently lower than 3%, mainly a result of the high market-value. Thus, if just strict economic considerations are taken into account, intervention thresholds lower than 2% could be considered. By contrasts, if environmental economic-injury levels are preferred [9] the 3% economic thresholds proposed would be a much better choice. In any case, these action levels should be optimally dynamic since variables for computation may strongly change over the cropping season or from site to site [3,10].

CONCLUSIONS

It may be concluded that larval density of H. armigera significantly affected both chickpea quality and yield. The higher the infestation level the higher the seasonal and harvest larval damages recorded and the lower the chickpea yield obtained. The implementation of the economic threshold has confirmed that damage should not exceed commercially acceptable levels if treatments follow the IPM scouting recommendations.

REFERENCES

1. Abott W.S., 1925. A method of computing the effectiveness of an insecticide. J. Econ. Entomol. 18, 265-267.

2. Anonymous, 2000b. Helicoverpa armigera Hubner. [in:] Crop Protection Compendium, 2nd Edition. (CD-ROM). CAB-International, Wallingford-Oxon.

3. Brown L.G., Hartstack A.W., Parvin Jr. D.W., Skieth R.W., 1979. Computer simulation for establishing economic thresholds. [in:] Sterling, W.L. (Ed.), Economic Thresholds and Sampling of Heliothis species on Cotton, Corn, Soybeans and other Host Plants. Southern Cooperative Series, Bulletin No. 231. Texas A & M University, Texas, pp.75-84.

4. Cameron P. J., Walker G.P., 1988. Bacillus thuringiensis for control of tomato fruitworm. Proc. N.Z. Weed Pest Control Conf. 41, 99-100.

5. Dent D., 1991. Insect Pest Management. CAB-International, pp. 127-131.

6. Fitt G.P., 1989. The ecology of Heliothis species in relation to agroecosystems. Annu. Rev. Entomol. 34, 17-52.

7. King E.G., Coleman R.J., 1989. Potential for biological control of Heliothis species. Annu. Rev. Entomol. 34, 53-75.

8. Pawar C.S., Bhatnagar V.S., Jadhav D.R., 1986. Heliothis species and their natural enemies with their potential for biological control. Proceedings of Indian Academy of Sciences (Animal Sciences) 95, 697-703.

9. Pedigo L.P., 1996. Entomology and Pest Management, 2nd Edition. Prentice-Hall, Englewood Cliffs, NJ.

10. Pitre H. N., Mistric W. J., Lincoln C.G., 1979. Economic thresholds: concepts and techniques. [in:] Sterling, W. L. (Ed.), Economic Thresholds and Sampling of Heliothis species on Cotton, Corn, Soybeans and other Host Plants. Southern Cooperative Series, Bulletin No. 231. Texas A & M University, Texas, pp. 12-30.

11. Raheja A.K., 1996. IPM Research and Development in India: Progress and Priorities. In Recent Advances in Indian Entomology (Ed. O.P.Lal). APC Publications Pvt. Ltd., New Delhi, pp. 115-126.

12. Sokal R. R., Rohlf F. J., 1981. Biometry, 2nd ed. Freeman, New York.

13. Systat, 1990. Systat: The System for Statistics. Systat Inc., Evaston, IL.

14. Torres-Vila L.M., Rodr-Molina M.C., Palo E., Del Estal P., Lacasa A., 2000a. El complejo parasitario larvario de Helicoverpa armigera Hubner sobretomate en las Vegas del Guadiana (Extremadura). Bol. San.Veg. Plagas 26, 323-333.

15. Torres-Vila L.M., Rodr-Molina M.C., Palo E., Bielza P., Lacasa A., 2000b. La resistencia a insecticidas de Helicoverpa armigera Hubner en Espana: datos disponibles. Bol. San. Veg. Plagas 26, 493-501.

16. Walker G. P., Cameron P. J., 1990. Pheromone trapping and FIeld scouting for tomato fruitworm in tomatoes and sweet corn. Proc. N. Z. Weed Pest Control Conf. 43, 17-20.

17. Zalom F. G., Wilson L. T., Smith R., 1983. Oviposition patterns by several lepidopterous pests on processing tomatoes in California. Environ. Entomol. 12, 33-1137.

18. Zalucki M. P., Daglish G., Firempong S., Twine P., 1986. The biology and ecology of Heliothis armigera (Hubner) and H. punctigera Wallengren (Lepidoptera: Noctuidae) in Australia: what do we know? Austr. J. Zool. 34, 779-814.

19. Zalucki M. P., 1991. Heliothis research methods and prospects Springer, New York. p. 181.

Accepted for print: 12.03.2007

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.