Effects of irrigated and rain fed conditions on infestation levels of thrips ( Thysanoptera : Thripidae ) infesting Dolichos lablab ( L . ) in Eastern Kenya

Dolichos lablab (L.) is a neglected crop species in Kenya, grown only for subsistence where they are highly adapted to a small range of ecological niches. It is grown under irrigation and rain fed conditions, where production is limited by insect pests such as thrips (Genera: Megalurothrips and Frankliniella). However, the effects of irrigation and rain fed conditions on infestation levels of these thrips in Kenya are unknown. This study aimed at investigating the levels of infestation of thrips on D. lablab (L.) grown under rain fed and irrigated conditions in Meru Central and Yatta sub-counties, Kenya. Sampling was conducted biweekly from June, 2009 to February, 2010. In each sub-county, eight farms (four each irrigation and rain-fed) were randomly selected. The infestation levels were monitored using 5 quadrats (30 cm × 30 cm), placed randomly at each farm in every sampling session. The thrips population within each quadrat was estimated from 3 randomly selected shoots of D. lablab. Thrips samples were placed in 30 ml vials and transported in a cool box to the International Centre for Insect Physiology and Ecology (ICIPE) labs for sorting and identification. The abundance of thrips varied significantly between the irrigated and rain fed farms at both sub-counties. In Meru sub-county, thrips numbers averaged 352.3±36.1 for irrigated farms as compared to 199.1±26.3 for rain fed. In Yatta subcounty, the mean abundance for rain fed farms was 265.5±42.0 as compared to 235.4±37.2 for irrigated farms. D. lablab for irrigated farms were significantly more infested (Anova: p=0.05) than in rain fed farms.


INTRODUCTION
Dolichos lablab (L.), also known as Lablab purpereus (L.) is a neglected and under cultivated legume crop in Kenya and other parts of the world (Williams and Haq, 2002), as it is grown only in local production systems, where they are adapted to a small range of ecological niches and are highly underdeveloped due to lack of formal research input (Padulosi et al., 2006).In Kenya, the main D. lablab producing areas are Eastern, Central and Coast regions, where it's grown either as a pure stand or as an intercrop with maize (Ministry of Agriculture, Kenya Government, 2005).
D. lablab is a drought resistant and highly nutritious crop, but production is severely limited by pest infestations, especially thrips.The few farmers growing the crop have had decreasing levels of attainable yields due to pest infestations, with production ranging from 800 to 900 kg ha -1 , which is far less than the estimated yield potential of 2700 to 3000 kg ha -1 (Kamau et al., 2010;Kinyua et al., 2008).As a result of the low production, D. lablab is often being replaced with other legume crops such as common and French beans, which may not withstand the harsh environmental conditions in such areas.
In Kenya, D. lablab is grown under both irrigation and rain fed conditions.In the eastern region, the vegetable type is grown under irrigation while the grain type relies on rains.The vegetable type is locally known as Varole and is a perennial type grown mainly for its young fresh pods.It is trellised and climb on the stakes up to three meters high before it starts growing horizontally.It remains ever green throughout the production period as long as there is water for irrigation.The grain type on the other hand is annual and mainly grown for its fresh or dry grains.It is drought resistant and remains ever green during the dry season long after other crops have dried (Maundu et al., 1999).In both sub-counties, rains have been reported to be erratic and unreliable with annual average ranging between 500 to 800 mm (Wambugu et al., 2010) The proliferation of insect pests such as thrips is more favorable in warm climates (Anonymous, 2006).In such climates, irrigation can stimulate the insect pest population levels (Kannan and Mohammed, 2001).In India, production of D. lablab is limited by a wide range of insect pests, but thrips were found to be the major insect pests in both irrigated and rain fed conditions (Thejaswi et al., 2007;Rekha and Mallapur, 2009).For D. lablab, irrigation can extend the growing seasons which enables insect pests to complete a greater number of reproductive cycles (Kannan and Mohammed, 2001).This can lead to increased pest infestations, with farmers needing to apply higher levels of chemical pesticides with an aim of controlling D. lablab pests, particularly thrips.
In Kenya, information on infestation levels of thrip species associated with D. lablab in irrigated and rain fed conditions is lacking.However, thrips of the genus Frankliniella and Megalurothrips have been reported to infest D. lablab in the eastern region of Kenya (Kamau et al., 2010).These genera were found in young leaves, flower buds, flowers and young pods.In India, rain fed and irrigated D. lablab, severe infestation was characterized by flower and pod malformation, distortion and scarification, while flower buds did not open but aborted prematurely (Thejaswi et al., 2007).
This study aimed at investigating the effects of irrigated and rain fed conditions on the thrips species and their infestation levels in D. lablab.

MATERIALS AND METHODS
Thrips infestation levels on D. lablab were evaluated biweekly throughout the two seasons in Yatta (37°53' E and 1°56'S) and Meru Central (37°30'E and 2°50'S) Sub-counties, in the eastern region of Kenya (Figure 1).In Yatta, rains were reported to be erratic and unreliable with annual average ranging between 500 to 800 mm (Wambugu et al., 2010) while in Meru, the annual average received was as low as 500 mm.The average lowest and highest temperatures in Meru were recorded as 16 and 23°C respectively (Oginosaka et al., 2006).
Average minimum and maximum temperatures for Yatta were 13.8 and 30.7°C sequentially (Kang'au, 2011).Studies were carried out under farmers' field conditions during the dry (June, 2009 to October, 2009) and wet seasons (November, 2009to February, 2010).In both sub-counties, D. lablab was grown under irrigation and rain fed conditions.Four-irrigated (vegetable type) and four rain-fed (grain type) farms were randomly selected in each subcounty, totaling 16 farms (farms 1 to 8 in Meru and 9 to 16 in Yatta) in the 2 sub-counties.At each farm, thrips infestation levels were assessed from a set of 4 randomly selected D. lablab plants; which were selected from five randomly selected points on the study farms.The sampling of D. lablab was done at the seedling and vegetative stages, before the twigs intertwined to form a single canopy during every sampling session.Infestation levels were further assessed at the vegetative and podding stages, when the crop twigs intertwined and formed a dense canopy using a 30 cm x 30 cm quadrat (0.09 m²).
From each set of four D. lablab plants or the plant parts within the quadrat, thrips populations were estimated by picking three randomly selected shoots, and their leaves, flowers and flower buds placed in 30 ml vials.The thrips samples were taken to ICIPE laboratories in Nairobi for sorting and identification.In the laboratory, the leaves, flowers and flower buds were opened and adult thrips and nymphs extracted by washing in 70% alcohol.All the thrips were then sorted, identified, counted and recorded.Identification was done using characteristics such as color, the rows of setae along the veins of the fore wings, the number of antennal segments and the positioning of the pair of setae III on the imaginary triangle of the occelli (Ralph, 1998;Stiller, 2001).
All the thrips species from the plant parts collected from the set of 4 plants or the quadrat were sorted according to their characteristics, identified, counted and recorded.The data was then analyzed using Genstat Statistical Software, version 14.To determine if the infestation levels of thrips on D. lablab differed significantly between irrigated and rain fed D. lablab, ANOVA was performed on transformed count data.Post hoc test was done using the Student-Neuman-Keuls test.The rejection level was set at α < 0.05.

RESULTS
Thrips species in irrigated and rain-fed D. lablab in Meru sub-county were dominated by Megalurothrips sjostedti during both dry and rainy seasons.During the dry season, adults of M. sjostedti infesting D. lablab were significantly higher (F (3,57) =4.73; p<0.05) on irrigated farms with a mean number of 176.2±11.0 as compared to 117.1±3.0 on rain fed farms (Table 1).Infestations were also significantly higher in irrigated farms (F (3,71) =8.41; p<0.05) during the wet season with a mean of 109.4±10.8 as compared to 19.6±7.9 thrips on rain-fed farms (Table 2).
In contrast, infestation rates by F. occidentallis in Yatta Sub-county were lower in the irrigated than in the rain fed D. lablab during the dry season (Table 1).However, Table 1.Means (±SE) of counts of various thrips species in irrigated (vegetable type) and rain fed (grain type) conditions in Meru central and Yatta sub-counties (Dry season).Means with the same letter within the same row are not significantly different at α<0.05.Means separated using SNK test.

DISCUSSION
From the results, thrips infestation levels were significantly higher in irrigated D. lablab compared to rain fed in Meru.Irrigation provided good growing conditions for the D. lablab plants to ensure rapid growth hence enough shoots and flowers for thrips to feed on.This is in line with the findings of Kasina et al. (2009) who found that flower thrips infestation was high in the shoots and flowers compared to other parts of the crop.In D. lablab, irrigation brings about longer growing seasons which enables these species of thrips to complete a greater number of reproductive cycles during the growing season (Kannan and Mohammed, 2001).However, environmental stress such as drought weakens plants and makes them more susceptible to thrips attack.This explains why the M. sjostedti infestation levels were higher in rain fed compared to irrigated in Yatta.This was observed by Gitonga (1999) in French beans who reported that plants under water stress are very susceptible to direct thrips damage.Megalurothrips sjostedti is widely spread in sub Saharan Africa and highly polyphagous in the family leguminosea (Seif et al., 2001).This suggests why M. sjostedti was more abundant than any other thrip species in this study.

Conclusion
Infestation levels of the species of thrips attacking D. lablab in Meru central and Yatta sub-counties of eastern region of Kenya varied significantly between irrigated and rain fed conditions.Megalurothrips sjostedti, F. occidentallis and F. schultzei infestation levels were significantly higher in irrigated D. lablab (vegetable type) than rain fed D lablab (grain type) in Meru central.This was extremely different in Yatta where M. sjostedti infestation levels were significantly higher in rain fed than irrigated.

Figure 1 .
Figure 1.Location of the study sites in Meru Central and Yatta Sub-counties, Kenya.

Table 2 .
Means (±SE) of counts of various thrips species in irrigated (vegetable type) and rain fed (grain type) conditions in Meru central and sub-counties Yatta (Wet season).Means with the same letter within the same row are not significantly different at p<0.05.Means separated using SNK test.