Influence of weather factors on the incidence of coffee berry borer, Hypothenemus hampei (Ferrari) (Scolytidae: Coleoptera) in Pulney hills, Tamil Nadu

: Studies were carried out to assess the influence of weather parameters on the incidence of coffee berry borer, Hypothenemus hampei (F.) at three locations in lower Pulney hills of Tamil Nadu during 2004 - 05 and 2005 - 06. In all the three locations, the infestation was higher (19.71%) in robusta than in arabica coffee varieties. In both the varieties, the peak period of infestation occurred between November and December. Correlation between weather parameters and infestation of coffee berry borer revealed a negative association with maximum temperature and rainfall at all the three locations while a positive relationship with maximum and minimum relative humidity with reference to the infestation of berry borer. Irrespective of the locations surveyed the population build-up of coffee berry borer in left over berries had served as a main source of inoculum for their carryover to the next season. The mean number of borer adults emerged from gleaning was high (21.72) due to rain followed by 12.93 and 12.52 recorded in mere water spray and surface temperature, respectively.


Introduction
The coffee berry borer, Hypothenemus hampei (Ferrari) (Scolytidae: Coleoptera) is one the most serious pests of coffee in many of the worlds chief coffee producing countries, which has caused great losses to the yield (Le-Pelley, 1968).Though, it gained entry in to India in 1990 via. Gudalur in Tamil Nadu it made its first appearance on Pulney Hills at Pethuparai village, an isolated pocket in Perumalmalai liaison zone (Anonymous, 1996).It invaded in to the main coffee growing areas of Pulneys during 1997 and by 2000 spread to entire coffee areas in Pulneys.The coffee berry borer is known to feed and reproduce only in the seeds of coffee species.The female beetle enters into the coffee berry by cutting a circular hole, generally at the tip of the berry.Occasional attempts made by the coffee berry borer to penetrate into the immature endosperm cause decaying of endosperm by secondary infection resulting in premature fruit drop.Any delay in harvesting will aggravate the damage as rate of reproduction is faster near to harvest (Baker, 1999;Sreedharan et al., 2001).In the recent years, the berry borer menace has been felt as a major limiting factor for quality coffee production by the coffee growers of lower Pulney hills.Information on the seasonal trends in population development of coffee berry borer is essential for timely implementation of cultural and chemical control methods.Hence, the present study was taken up on the influence of weather factors on the incidence of berry borer, H. hampei at three locations in Pulney hills, Tamil Nadu and results are presented here.

Seasonal infestation of coffee berry borer
A study was conducted in lower Pulney hills, to assess the seasonal incidence and influence of coffee berry borer, H. hampei at fortnightly interval from January 2004 to December 2005 in the coffee plantations (C.arabica and C. canephora) maintained at Regional Coffee Research Station, Thandigudi (4300 feet MSL), VKV hill garden estate, Periyamalai (3600 feet MSL) and Pillaveli estate (2900 feet MSL).In each plantation, observations were made from 10 randomly selected coffee plants and in each plant, three branches were chosen.The damaged berries by berry borer was worked out based on number of infested and uninfested berries.The data were corrected by using Abbot's formula.The mean were compared using Duncan's multiple range test (DMRT) (Erwin et al., 1962).The data on weather parameters, maximum and minimum temperature (°C), relative humidity (%) and rainfall (mm) were recorded during study period.

Source of inoculum of coffee berry borer in left-over berries
The left-over berries were collected from 17 locations on lower Pulney hills during May and June of 2004 and 2005.One hundred left-over berries were collected randomly in five places in each location and they were sliced and recorded for the total number of beetles per berry.

Adult emergence pattern of coffee berry borer from gleanings
The infested gleanings were collected from the field in six treatments and five replications after the main harvest.Fifty fruits were collected and maintained uniformly in each replication.The treatments were imposed with (1) water spray, (2) water soaking for 2 min., ( 3) exposure for natural rain, (4) exposure to surface temperature (25 °C), ( 5) impact of relative humidity (90%) and ( 6) untreated check.The above treated gleanings were kept in a plastic container covered with brass wire mesh on top to allow the aeration.The emerging adults were counted periodically up to 5 days.

Statistical analyses
The fortnightly means were calculated from the data on the incidence of coffee berry borer and this was correlated with the weather parameters.Data collected in various field and laboratory experiments were statistically analysed.The percentage values were subjected to arc sine transformation.Square root transformation was followed for converting the population / numbers.The treatment means were compared with Duncan's multiple range test (DMRT) for their significance (Gomez and Gomez, 1985).

Seasonal infestation of coffee berry borer
Seasonal occurrence and influence of weather parameters on the infestation of coffee berry borer, H. hampei in coffee plantation (C.arabica and C. canephora) during the period from January 2004 to December 2005 at RCRS (Farm), NTN estate (Thandigudi), VKV hill garden estate (Periyamalai) and Pillaveli estate (Pillaveli) of lower Pulney hills revealed the following.1).
Correlations between the weather parameters and the coffee berry borer infestation in C. arabica during 2004 and 2005 showed that maximum temperature (r = -0.411and r = -0.537)and rainfall (r = -0.507and r = -0.327)exhibited significant negative association with berry borer infestation while maximum relative humidity (r = 0.506 and r = 0.021) and minimum relative humidity (r = 0.629 and r = 0.619) had a positive association, (Table 2) respectively.
The correlation studies made between weather parameters and berry borer infestation in C. canephora during 2004 and 2005 revealed that maximum temperature and rainfall had significant negative association recording the r values of-0.302 and -0.497 in 2004 and -0.526 and -0.347 in 2005, respectively.Where as maximum and minimum relative humidity had a positive association with berry borer infestation (r = 0.529 and 0.172; r = 0.685 and 0.717 respectively.

Periyamalai
In Periyamalai, VKV estate, the peak per cent berry damage by berry borer in arabica was 13.17 and 9.17 per cent in first fortnight of November 2004 and 2005, while the least damage was recorded during March 2004 (0.02%) and 2005 (0.01%) respectively (Table 1).As far as robusta was concerned the highest infestation was recorded (17.11 and 14.17%) 3).The results also predicted that an increase in maximum temperature by 1 °C and rainfall by 1 mm would decrease the berry borer infestation by 1.51 and 0.53 per cent in 2004, and 1.23 and 0.11 per cent in 2005 respectively.But an increase of one per cent in case of maximum and minimum relative humidity increased the berry borer infestation by 0.44 and 0.25 per cent during 2004 and 0.07 and 0.17 per cent during 2005 in C. arabica.Minimum temperature had no significant correlation with the level of berry borer infestation during 2004 mand 2005.
During 2004, the maximum and minimum temperatures exhibited no significant association with berry borer damage in C. canephora.However, maximum (r = 0.479) and minimum (r = 0.600) relative humidity recorded significant positive association while the rainfall (r = -0.485)had a negative association with berry borer infestation.With reference to maximum temperature and rainfall during 2005, C. canephora showed significant negative correlation with berry borer infestation, with the 'r' values of-0.634 and -0.244 respectively, and minimum temperature (r = 0.175) and minimum relative humidity (r = 0.639) had positive influence on berry borer infestation (Table 3).

Pillaveli
In Pillaveli, in the different periods of observations, the per cent berry borer damage in arabica ranged from 0 to 17.15 per cent during 2004 and 0 to 10.17 per cent during 2005 (Table 1).The peak per cent damage by berry borer was 17.15 in second fortnight of November 2004, while it was 10.17 per cent in first fortnight of November 2005.In robusta, the per cent berry borer damage was maximum in second fortnight of November 2004 (19.71%) and first fortnight of December 2005 (17.17%).In above locations the infestation of berry borer was more in robusta than arabica.
The correlation co-efficient worked out between the berry borer infestation and weather factors in C. arabica during 2004 and 2005 indicated that maximum temperature (r = -0.205and r = -0.579)and rainfall (r = -0.297and r = -0.138)had negative correlation while maximum relative humidity exhibited positive correlation with the r values of 0.553 and 0.563 during 2004 and 2005, respectively (Table 4).
Simple correlation worked out between minimum temperature, maximum and minimum relative humidity and berry borer infestation in C. canephora during 2004 had shown no significant associations between them.
However, maximum temperature and rainfall recorded negative associations with the berry borer infestation recording the 'r' values of-0.174 and -0.476 respectively.During 2005 in C. canephora, the maximum temperature and rainfall exhibited its negative associations with berry borer infestation recording the r values of-0.542 and -0.176 and also predicted that an increase in maximum temperature by 1°C and one mm of rainfall would decrease the berry borer infestation by 0.99 and 0.09 per cent (Table 4).
Coffee berry borer was found to be the major pest of coffee at all the three locations surveyed in lower Pulney hills, Dindigul District, Tamil Nadu.Though its distribution had already been reported in coffee growing areas of Nilgris district, lower Pulney hills of Dindigul district, Wayanad district of Kerala and Kodagu district of Karanataka (Kumar et al., 1990 andSreedharan et al., 1994), the present investigation further quantified its occurrence in three major coffee growing locations on Pulney hills viz., Thandigudi, Periyamalai, and Pillaveli.During the fixed plot survey, observations were made on each plant from waist high plagiotropic branch and the number of infested and uninfested berries were recorded as stated by Baker et al. (1989).In all the three locations, the infestation was high (19.71%) in robusta than in arabica variety which is in conformation with the findings of Campos and Garcia (1997) who observed high percentage of infestation in robusta coffee (45 to 60%) than catimore, an arabica variety (32 to 41%) and it might be due to relative feeding preference to berries in robusta with prolonged fruiting season that favoured the infestation to a considerable level as reported by Baker et al., (1989).In both the varieties the peak period of infestation was noticed between November and December during which most of the berries were in ripening stage that preferred by coffee berry borer than earlier stages (Fig, 1 & 2).This finding gains support from the report of lboekwe (1984) who stated that coffee berry borer adults significantly preferred red coffee berries than green ones and Gaviria et al. (1995) reported that highest level of infestation occurred between 133 -220 days after flowering.In Pillaveli, the infestation was slightly higher than Periyamalai and Thandigudi, and this might be due to fluctuations in weather factors, shade effect, indiscriminate use of insecticides and by accumulation of left over berries infected with berry borer.
In all the three locations, infestation of coffee berry borer was gradually declined from January because most of the ripened berries were harvested during this period and thus the borers might be moved to over ripened and left over berries or dry berries (gleanings) for both shelter and further breeding and multiplication.This is in consonance with the earlier findings of ManSingh (1991) and Baker and Barrera (1993) who stated that prevalence of immature stages throughout the year and large number of females accumulated in the fallen berries during dry season.
The present investigation quantified its occurrence in three major locations viz., Thandigudi, Periyamalai and Pillaveli, during 2004 and 2005, and revealed that the coffee berry borer infestation in C. arabica and C. canephora had significant negative correlations with maximum temperature and rainfall while as positive relationship with maximum and minimum relative humidity.Similar results were obtained by Baker et al. (1992a) where increase in emergence of coffee berry borer at 90 -100 % RH in 20 -25 o C temperature regime which is in consonance with the present finding that a range of relative humidity ranged from 55.25 to 99.00 per cent and temperature regimes ranged between 10.25 and 32.75°C.Rainfall was possibly the factor that attributed to the low berry borer population (Ferreira et al., 2000).Further, Vijayalakshmi (2000) reported that the coffee berry borer infestation had a significant negative correlation with rainfall and found that only a thin population of the borer prevalent during the rainy months.This may be due to the mortality caused by the heavy rains.Earlier, Rehiman and Vijayalakshmi (1998) have also reported that rains cause mortality of the beetle.

Source of inoculum of coffee berry borer in left-over berries
The mean populations of coffee berry borer in different locations are presented in Table 5.The over all mean population of borer per gleaning recorded in Adalur, Solaikadu, Nallurkadu and Pillaveli was high as 38.16 ± 30.63, 36.16 ± 30.47, 35.74 ± 26.65 and 31.94 + 24.17 as against the less borer population recorded in Thadiyankudisai (23.62 + 21.54),Thandigudi (24.23 ± 19.19),Perumparai (24.26 ± 20.88) and Managalamkombu (24.83 + 21.00).Thus, irrespective of the locations surveyed the population build up of coffee berry borer in left over berries had become a main source of inoculum for their carryover to next season crops.

Adult emergence pattern of coffee berry borer from gleanings
Emergence of coffee berry borer from gleanings exposed to different treatments up to 5 days after collection and their results are presented in Table 6.The mean number of borer adults emerged from gleaning was high (21.72)due to rain followed by 12.93 and 12.52 recorded in mere water spray and surface temperature, respectively.Atwal and Balraj Singh (1990) reported that berry borer when exposed to extreme temperature might become dormant and resume activity on being exposed to favourable range.
Emergence of coffee berry borer from gleanings showed that maximum emergence of borer when the samples were exposed under natural shower and minimum emergence at 90% relative humidity.This is in confirmation with the earlier results that heavy rain also triggered the emergence of the beetles (Sreedharan et al., 1994) and low humidity (< 60% RH, 25 °C) provoked rapid evacuation of adults and while it was minimum at 90% RH (Baker et al., 1992a) during second fortnight of November 2004 and first fortnight of December 2005 and it was the least (4.08 and 4.09%) during 2004 and 2005 respectively.The simple correlations worked out between weather parameters and berry borer infestation in C. arabica during 2004 and 2005 indicated significant negative association of the maximum temperature and rainfall with berry borer infestation, recording the 'r' values of -0.559 and -0.535 during 2004 and -0.597 and -0.124 during 2005 respectively, and the maximum (r = 0.565 and r = 0.057) and minimum relative humidity (r = 0.695 and r = 0.654) exhibited positive associations with berry borer infestation during 2004 and 2005 (Table

Fig. 1 .
Fig.1.Seasonal infestation of coffee berry borer at three different locations