Distribution and impact of Busseola fusca (Fuller) (Lepidoptera: Noctuidae) and Chilo partellus (Swinhoe) (Lepidoptera: Crambidae) in Northeastern Ethiopia

An assessment was done during the long and short rainy cropping seasons of 2010/2011 across different agro-climatic zones (ACZs) with the objective of determining the distribution, species composition and damage levels of important stem borer species in North-eastern Ethiopia. Data were collected from 21 localities of six districts, four fields in each locality, and a total of 84 fields were assessed. Busseola fusca composed of 16 to 100% of the population of stem borer species and caused damage levels of 3 to 60% in South Wollo zone at the elevation ranging from 1750 to 2338 m. Moreover, Chilo partellus is composed of 7 to 100% of the population of stem borer species and cause a damage of 1 to 100% in the same zone at the elevation range of 1492 to 2084 m. Similarly, in North Wollo zone, B. fusca shared 69 to 88% of the total population of stem borer species and cause damage levels of 5 to 53% of the elevation ranging from 1850 to 2044 m, while C. partellus composed of 12 to 31% and caused damage of 2 to 26%. In the Oromia administrative zone, which falls at the elevation of 1400 to 1669 m, 100% of the stem borer population was C. partellus and caused a damage level of 84 to 99%. The result indicates that C. partellus widened its distribution and extent of damage from the previous report of 1900 up to 2044 m. The two stem borers were found in elevation ranging from 1750 to 2044 m but their level of distribution, compositions and damage varied between elevations. Conclusively, C. partellus widened its distribution and might have replaced the indigenous species, because it was recorded as up to 2044 m, which was not reported in the previous decades.


INTRODUCTION
Lepidopteran stem borers are generally considered to be the most damaging insect pests of maize and sorghum in Africa (Seshu, 1998).Maes (1998) reported 20 economically important stem borer species whose distribution, relative abundance and pest status are expected to vary with environmental conditions.Busseola fusca and the exotic Chilo partellus are the dominant species in East Africa.C. partellus has proven to be a highly competitive colonizer in many areas of eastern and southern Africa, often becoming the most injurious stem borer and displacing native species (Kfir, 1997).Studies in coastal Kenya showed that C. partellus has partially displaced the indigenous stem borer, Chilo orichalcociliellus (Lepidoptera: Crambidae) (Ofomata et al., 1999).In the Eastern province of Kenya, C. partellus although present in the early 1980s, it was less abundant than B. fusca (Seshu, 1983).However, in the same area in the period 1996 to 1998, B. fusca became rare and C. partellus became dominant (Songa, 1999).Similarly, in the eastern highveld region of South Africa, C. partellus partially displaced B. fusca over a period of seven years (Kfir, 1997).The displacement was most evident in grain sorghum where the proportion of C. partellus in the total stem borer population increased from about 3% in 1986 to 1991% in 1992 (Kfir, 1997).
Since the appearance of C. partellus in African continent in 1932, it has continuously expanded its distribution in the warm, low-altitude regions of eastern and southern Africa (Kfir, 1997).The same author reported that C. partellus was expanding its distribution into the high elevations of the eastern highveld region of South Africa.The only stem borer already found at an elevation of 1,600 m was B. fusca.After C. partellus invaded this region, it rapidly increased its share of the total borer population every year.In maize, it reached 32% of the total borer population within six years and on grain sorghum, 59% within seven years.Within two years, it became the predominant borer; constituting 90% of the total stemborer population.One of the possible reasons for the replacement of the indigenous species is hibernating for larval populations of C. partellus terminate diapause and emerge in one month earlier than B. fusca.This enables C. partellus to infest the grain sorghum before B. fusca, thus becoming the predominant borer in this niche.In addition, the life cycle of C. partellus is three weeks shorter than that of B. fusca, which gives it a further competitive advantage because of its higher potential rate of an increase (Kfir, 1997).Emana et al. (2008) reported from his survey that B. fusca was the dominant stem borer species in high-potential zones (highland tropics, moist transitional zone and moist mid-altitude) while the exotic C. partellus dominated smallholder farms in low potential zones (dry mid altitude, dry transitional and lowland tropical zone).In India, Gupta et al. (2010) reported that C. partellus, one of the most destructive pests of maize and sorghum, is the most important at altitudes below 1500 m above sea level.
In Ethiopia, B. fusca and C. partellus are considered to be the most damaging insect pests, with reported yield losses of 0 to 100, 39 to 100, 10 to 19 and 2 to 27% from South, North, East and Western Ethiopia, respectively (Melaku and Gashawbeza, 1993;Melaku et al., 2006).
Previous two decades, Assefa (1985) reported that C. partellus was a predominant species at lower elevation of less than 1700 m and B. fusca was dominant at high elevation of 1160 -2600 m.a.s.l. and in cooler areas.Emana et al. (2001) conducted a survey in 1999 and 2000 and reported that C. partellus widened its distribution from 500 -1700 to 1030 -1900 m.a.s.l.whereas B. fusca was recorded between 1030 -2320 m.a.s.l.However, studies were conducted in different on the compositions, distribution and damage levels of these stemborer species in northeastern Ethiopia.Thus, understanding species distribution and abundance of stem borer communities will constitute basic information necessary for future development of management strategies.Therefore, the objective of this study was to know species composition, levels of damage and status of the most important stem borer species at different agro-ecological zones of northeastern part of Ethiopia.
Rain fed sorghum and irrigated maize is practiced in all districts.The study areas experience bimodal rainfall, the short rainfall in April or May and main rain June to September.Late and early maturing sorghum cultivars are planted in the short and main rainfall seasons, respectively.During the study period, all fields were covered with late maturing sorghum cultivar, Degalit.Based on the classification of agro-ecological zones (AEZs), almost all the zones experienced sub-moist warm (Kola, <1,500 m), dry-warm to moist cool (Woinadega, 1,500 -2,500 m) and moist cool (Dega, 2,500 -3,500 m) (MoA, 2000) (Table 1).

Sampling procedures
Sampling procedures were done following the work of Emana (2001Emana ( , 2008)).Randomly, selected 3 to 4 peasant associations (PAs) were considered from each district.Four well managed farmers' fields (chemical untreated) with a minimum size of one hectare were selected.The fields were selected on the basis of accessibility and area coverage of sorghum and maize.A total of 84 fields from 21 PAs was assessed (Table 1).In each field, five plots with 3 × 3 m sizes were sampled in 'Z' fashion.
The assessments were conducted in seedling, booting and harvesting stages of sorghum and at tasseling of irrigated maize.The same fields were used for different time of sampling.The presence of the two stemborer species in the same field was confirmed at the seedling stage of sorghum.The presence of the two stemborers in the same field at seedling stage was confirmed using their oviposition sites on the plants.Egg oviposition sites of the two species are on different parts sorghum.Female B. fusca laid eggs between the leaf sheath and stem whereas C. partellus lay eggs on the undersides of leaves, mainly near the midribs.In fields where both the stem borers were found, all the plants in the plots were counted, dissected and data on the number of plants with leaf damage and larval density were collected.It was rare to find the two species on the same plant.In fields where a single species was found, six plants per plot were dissected and similar types of data were collected.At harvest, in each plot the total number of stems, and damaged stems were counted.The severity of leaf damage on sorghum/maize due to the two stem borer species was based on the assessment of the amount of feeding by the insects on the upper leaves in each plot at seedling (small size larvae), booting (medium size larvae) and harvest (large size larvae) stage of sorghum and at tasseling stage of maize.The leaf feeding damage parts of the country, there are no exhaustive information was scored on a 0 -5 scale, where 0 = 0 (no damage symptom), 1 = 1 -20%; Map 1. Map of northeastern Ethiopia showing localities surveyed in the three zones.2 = 21-40%; 3 = 41-60%; 4 = 61-80% and 5 = over 81% damage or complete death of the plant (Kalule et al., 1997).
The proportion of infested plants (IP) was calculated in relation to total sampled plants (TP).IP (%) = , where, IP = infested plants, TP = total plants in a quadrant.The larvae and adults of B. fusca and C. partellus were identified using identification keys, color pictures and morphological differences.C. partellus larvae have a cream to pink coloration, with dark spots along the dorsal surface; the head capsule is brown.Moreover, C. partellus larvae can be distinguished from B. fusca by the presence of a complete circle crochets on the prolegs whereas in B. fusca the crochets are arranged in a crescent (Hutchison et al., 2008).The field identification was confirmed by rearing fifty of the identified larvae per plot to the adult stage in the laboratory.Adult moth identification was carried out.The selected maize fields were sown from December to January using irrigation and sorghum in April/May using short rain period.In all districts, farmers practiced sowing maize under irrigation.Fields in all the surveyed districts were covered by Degalit and all the data in this study were from this cultivar.Data on temperature (maximum and minimum), rainfall (mm) and relative humidity (%), of each locality were obtained from the nearby meteorological stations.However, few stations do not have data on relative humidity, and they call it third class stations (Table 1).Coordinates and elevations of each field were recorded using global positioning system (GPS) 2000-2007 with brand name Garmin Ltd or its subsidiaries.

Data analysis
The data were arranged in nested design and analyzed using SPSS Version 12 software.The significantly different means (<0.05) were separated using Student-Newman-Keuls (SNK) multiple range test (Gomez and Gomez, 1984;Hinkelmann and Kempthorne, 2008).To normalize the data, Arcsin and square root transformations were used for percentage and count data, respectively, and distrasformed thereafter.Correlations of different data were also analyzed.
Percent damage and number of B. fusca larvae had weak positive correlation with elevation with r = 0.2, P = 0.663 and r = 0.1, P = 0.083, respectively.However, percent damage and number of C. partellus larvae had significant high inverse relation with elevation with r = -0.76,P<0.001 and r = -0.71,P<0.002, respectively.Moreover, percent damage and the number of larvae had positive correlation with r = 0.82, P <0.001 (B.fusca) and r = 0.89, P < 0.001 (C.partellus).In addition, temperature and composition of C. partellus and B. fusca had positive       Bf = Busseola fusca, Cp = Chilo partellus, mean (± SE) within columns, along each district, followed by the same letters do not differ significantly at the 5% threshold (Student-Newma n-Keuls-SNK).
contrast, percent damage and number of larvae of C. partellus had significant high inverse relationship with elevation with r = -0.92,P < 0.001 and r = -0.83,P<0.001, respectively.Similarly, percent damage and number of larvae of B. fusca and C. partellus were positively correlated with r = 0.58, P = 0.019 and r = 0.83, P < 0.001, respectively.In addition, temperature and composition of C. partellus and B. fusca had positive and significant negative correlations with r = 0.83, P = 0.170 and r = -0.91,P = 0.020, respectively.
In this zone, C. partellus was the most economically important insect that caused severe damage on maize and sorghum.

DISCUSSION
Composition, damage levels and densities of B. fusca and C. partellus on maize at tasseling and sorghum at different growth stages were varied between elevations of different districts of the three administration zones.In all zones, longer maturing sorghum growth stages were synchro-nized with different rainfall periods, the seedling stage in short rainfall period and the booting stage in main rainfall period.
In Tehulederi district, larval density of B. fusca was varied between sorghum growth stages and the highest density was at booting and the lowest at seedling stage.The reason could be that the first generation of this insect occurred in late April after a long larva diapause period during short rainfall.The long larval diapause period influenced the number of eggs laid and the survival of neonate larvae which depend on the food reservoir of diapause larvae and the second reasons could be the abiotic and biotic factors which reduce the population at each pest stage.
In contrast, the second generation occurred in early July (main rain period) after a short larval diapause period that caused a high number of eggs and high survival of neonate larvae.Hence, higher numbers of larvae were recorded at booting stage than at the seedling stage of sorghum.The results also concurred with Kioko et al. (1995).However, because of short larval diapause period and greater than two generations during short rainy period, higher C. partellus densities were recorded at seedling than at booting growth stages.Similar results were reported by Bernays et al. (1983), Ampofo and Kidiavai (1987), Kumar (1992), Kumar et al. (1993) and Odindo et al. (1992).With only one cropping season a year, followed by the long and dry off-season, C. partellus should not become a serious problem.
The composition of B. fusca varied from 51 to 73% in Habru district (1850-1889 m), 69-88% at Gubalafto (1856-2004 m), 65 to 100% at Tehulederi (1887-2338 m) and 58-93% at Kalu (1834-2084 m) with mean minimum and maximum temperatures of 9.6-13.7 and 19.5-26°C, respectively.This species was not recorded at elevations less than 1770 m, but it shared 16% and caused 2.5% sorghum damage in the 1770 m at Jari with maximum temperature of 27.2°C.C. partellus was recorded in all localities where they are situated at elevations between 1492-2084 m.The proportion of C. partellus increased from 7 to 100% following the drop of elevation from 2084 to 1492 m and with the increment of temperature from 24.6 to >28.3°C.The result of this study shows that distribution of B. fusca had a positive relationship with elevation and rainfall, and an inverse relationship with temperature.
However, distribution of C. partellus had inverse relations with elevation and rainfall but a significant positive relationship with temperature.Although, the rainfall, temperature, relative humidity and topographic features are different in different countries of Africa, these results partly in agreement with different authors that B. fusca occurs in West Africa from sea level to altitudes in excess of 2000 m but is most abundant in the wetter parts of the tree savannah in Burkina Faso (Nwanze, 1988).In eastern Africa, it occurs between 600 and 2700 m and the species is unable to tolerate mean temperatures of above 27°C (Harris and Nwanze, 1992).Sithole (1989) reported that B. fusca is the dominant stem borer at elevations above 900 m and indicating the ability of this pest to adapt to low-lying and the warmer area of South Africa.Tsedeke et al. (2000) also reported that the distributions of B. fusca and C. partellus are affected by rainfall, temperature and elevation in sub Saharan countries.
In Ethiopia, Emana (2001) reported that C. partellus was recorded in elevation between 1030 to 1900 m.In Pakistan, Muhammad et al. (2010) reported that the highest infestation of C. partellus was found at a temperature of 32.5°C, low annual rainfall (<1000 mm) and relative humidity of 68%.In Eritrea, it was reported that B. fusca and C. partellus were important in highlands between 1450-2350 m and low lands less than 1400 m, respectively (Adugna and Hofsvang, 2001).But in contrast to other authors, this result confirmed that B. fusca was not recorded in elevation less than 1750 and C. partellus was extending its distribution from 1400 to 2084 m.That means it is not only the elevation that influence the distribution of the two species but also other factors like rainfall, temperature, relative humidity and topographic features plays a great role (Nwanze, 1988;).The elevation at which we recorded C. partellus is higher than the previous record in Africa which was between 1700-1900m (Seshu, 1983) ) and less than 1700 m Assefa (1985).
This result showed that C. partellus expanded its ecological niche from 1700 m to >2000 m.Moreover, these survey results show a discrepancy from a previous survey report of the Sirinka Agriculture Research Center (SARC) in 1995 that revealed B. fusca was the only lepidopterous species attacking sorghum and maize at Sirinka PA and Gubalafto district with an elevation of 1850-1889 and 1808-2044 m, respectively (Adane and Tesfaye (1998), personal communication).
However, after 15 years of SARC report, this comprehensive survey was conducted and confirmed that C. partellus shared 12 to 49% of the stem borer population in these areas and expanded its ecological niche up to 2044 m.This showed that this invasive exotic species coexists with the native indigenous species, B. fusca, and may totally replace it in the near future.Similar to the findings of this research, different authors also reported that C. partellus could co-exists in many areas with B. fusca and expand its ecological niche from lowlands to highlands (Polaszek, 1998).Kfir (1997) reported that B. fusca was the only stem borer found in elevation 1,600 m in South Africa but after C. partellus invaded the region, it reached 32% of the total borer population of maize and 59% of sorghum within 5-7 years.The same author reported that C. partellus has proven to be an efficient colonizer, and it seems to be displacing the indigenous B. fusca.The only species that accounted for 100% of the two stem borer population in elevation less than 1700 m was C. partellus.Regardless of crop types and stages, sorghum damage of 70 to 96% and maize damage of 70 to 91% by C. partellus was recorded at elevations between 1419-1670 m.Similarly, Gupta et al. (2010) reported that damage magnitude of C. partellus ranges from 26.7 to 80.4% in altitude below 1500 m in India.However, damage level of C. partellus was as low as 5-21% in elevation between 1850-2044 m.In line with these results, various reports indicated that C. partellus distribution is highly influenced by altitude and moisture gradients.In Kenya, for example, C. partellus populations are most common in the dry mid altitude and dry coastal areas, but the pest also occurs in the moist-transitional and moist mid-altitude (<1500 m) agro-ecological zones (De Groote et al., 2003).
Our study indicates that the distribution of the two stemborer species are limited by elevations and the data also suggests a trend for increased infestation from high potential to low potential localities, namely the highlands (low infestation) and the lowlands (high infestation).Results from the presented studies provide an increased understanding on the ecology of these stemborer species which have meaningful positive impacts in the population dynamics and in designing effective, economical and safe integrated control measures especially host specific biological control agent like Cotesia flavipes, the one parasitizing C. partellus is not likely the parasite of B. fusca.Moreover, the information is useful in monitoring and forecasting the two species.

Table 3 .
Mean (-SE) damage score (1-5 scale) of the two stemborer species on sorghum and maize South and North Wollo zones in 2010/11.

Table 1 .
Elevation (m), temperature (°C), relative humidity (%) and rainfall of each locality of the three zones of northeast Ethiopia in 2010.
na = data not available.

Table 2 .
Mean (± SE) infestation (%) of maize and sorghum due to the two stem borer species in South and North and Wollo zones in 2010/11.