Introduction

Many North American insects, including several that are important pests, undertake long-distance migrations (Table 2.1). Knowledge of the role of migration in the population processes of these species, and of the relation of migration to climate and atmospheric phenomena, has developed in four broad phases. At first it was simply recognised that serious pest problems in the northern latitudes of the USA were sometimes initiated by populations that appeared suddenly in association with southerly winds. Then in the 1920s and 1930s, airplane sampling of the upper air became possible (Felt, 1928, 1937; Glick, 1939) and it was found that insects, spiders, and mites were commonly to be found at heights up to 1.5 km and occasionally (<1% of specimens obtained) at greater altitudes, even up to ~4.6 km. Over 700 species were caught in these airplane samples, but certainly not all were long-distance migrants. Seasonality, light intensity, temperature, wind speed and direction, and some other meteorological variables were found to be important factors affecting the distribution of insects in the upper air. In the third phase of migration research, during the 1940s, 1950s and 1960s, the association between the sudden appearance of a species in an area and the passage of synoptic-scale (200–2000 km) weather systems was documented in a series of studies.

Introduction

The long-range migrations of the the Brown Planthopper Nilaparvata lugens and the White-backed Planthopper Sogatella furcifera extend over the whole area of rice cultivation in East Asia and are closely associated with the wind systems of the monsoon climate of the region (Chapter 3, this volume). Because they do not overwinter in Japan, field populations of these planthoppers originate with overseas immigration in the Bai-u (rainy) season in June and July.

Kisimoto (1976) found that immigration of rice planthoppers into Japan was associated with the activity of frontal depressions during the Bai-u season and concluded that mass immigration was correlated with depressions that had passed through eastern central China. The southwest winds in these depressions carry the migrating planthoppers and Rosenberg & Magor (1983) used trajectory analysis to estimate the source of N. lugens arriving in Japan or caught on board ship in the East China Sea. They showed that trajectories based on winds at 1.5 km above sea level originated in potential emigration source areas more frequently than trajectories at 10 m.

Seino et al. (1987) obtained highly significant correlations between the immigration of planthoppers into Japan and the development of lowlevel jet (LLJ) airflows (see below) that coincide with the occurrence of heavy Bai-u rainfall. This chapter discusses the use of analysis of LLJs in monitoring systems to determine the location and timing of rice-planthopper immigration into Japan.

Introduction

The Oriental Armyworm Mythimna separata is a serious pest of gramineous crops and pastures in Asia and New Zealand (Sharma & Davies, 1983). In Japan, outbreaks occur every few years, mainly along the western (Sea of Japan) coasts of Hokkaido and Honshu and in western Kyushu (Fig. 6.1). Outbreaks affecting 200 ha or more of crops occurred in eight of the 31 years 1958–1988: in I960,1971, 1972, 1978, 1981, 1982, 1984 and 1987. The most serious outbreaks (over 20 000 ha affected) were in 1960, 1971 and 1987 (Oku & Kobayashi, 1977; Hirai, 1988a). These outbreaks are of significant economic importance and some required large-scale insecticidal control.

Since 1941, insect pests of rice (especially planthoppers (Delphacidae) and leafhoppers (Deltocephalidae)) have been monitored in Japan with a nationally operated light-trap network. The network extends throughout the country, with four or five traps in each of the 46 prefectures. M. separata is caught in these traps, but only in small numbers: the incandescent lamps used do not attract this species efficiently. A smaller network of food-lure traps (containing a mixture of sake lees, water, brown sugar and vinegar in the ratios 13 : 32 : 5 : 1 by weight) has therefore also been set up specifically to monitor M. separata. It extends to just four prefectures: Hokkaido, Aomori and Akita in the north and Kagoshima in the south.

Introduction

There is now strong evidence that the arid zones of Australia, the driest continent on Earth, play a vital role in the annual life cycle of several of the country's most significant insect pests. This chapter provides an overview of recent findings on the migratory behaviour of the Common Army worm Mythimna convecta, and of how it leads to outbreaks in the agricultural zones following population buildup in the dry inland. The following two chapters provide similar treatments for the Native Budworm Helicoverpa punctigera and the Australian Plague Locust Chortoicetes terminifera.

Australia's climate is characterised by relatively low and irregular rainfall: more than 80% of the continent receives less than 600 mm of rain annually (Fig. 7.1a,b). There are major interannual fluctuations in rainfall across the continent which are attributable to the El Niño / Southern Oscillation (ENSO) phenomenon (Nicholls, 1991). The most predictable rainfalls occur in the coastal areas, and regularity and quantity diminish rapidly inland (Division of National Mapping, 1982). The southern part of the continent receives most of its rain in winter in conditions associated with mid-latitude frontal depressions whose centres move eastwards at a latitude of about 40° S. The tropical and subtropical climate zones cover the northern half of Australia and are characterised by rainfall occurring mainly in summer. However, much of the inland of the continent is arid. It is neither far enough north to receive regular monsoon rainfall, nor sufficiently far south to be often visited by rainbearing cold fronts.

Western Australia: an environment for migration?

The state of Western Australia (WA) occupies approximately one-third of the Australian continent. It consists principally of a gently undulating plateau 300–600 m above sea level; there are no outstanding mountain ranges. Three broad climatic zones are recognised (Fig. 9.1): a tropical region in the north, with 90% of annual rainfall (500–1200 mm) falling between November and March (summer); a Mediterranean region in the southwest with 80% of annual rainfall (400–1200 mm) falling between April and September (autumn, winter and early spring); and an arid (desert and semidesert) zone, occupying the great majority of the state's land area, in which rainfall averages less than 300 mm per year. Cropping is almost completely confined to the Mediterranean region, and occurs principally in a broad band towards its boundary with the arid zone. Within the vast arid zone, rainfall occurs mainly during summer in the north, during both summer and winter in the centre, and during winter in the southwest. It is non-seasonal in the southeast (Fig. 9.1b). Rain usually falls on only a few days each year and drought conditions can persist for months or even several years. Temperatures are generally high, with mean monthly maxima exceeding 37 °C in the hottest 3–5 months. Average annual evaporation is more than ten times the rainfall.

Introduction

The Oriental Armyworm Mythimna separata is a severe pest of cereal crops, especially wheat, maize, millet and rice, throughout eastern China. Historical records suggest that the species has been a pest on millet and wheat for thousands of years (Zou, 1956). In the northeastern provinces of Liaoning, Jilin and Heilongjiang (Fig. 4.1), where the climate is highly seasonal, with severe winters and a growing season lasting only from April to September, an initial generation of M. separata larvae regularly causes severe damage to spring wheat, millet and maize in July, and a second generation damages millet and maize, and sometimes also rice, in August. The initial infestation in the northeast arises from egg-lays by moths that appear every year, often in large numbers, in late May and early June (Chen, 1962). An intensive and comprehensive programme of research on this key pest of Chinese agriculture during the late 1950s and early 1960s prompted the proposal, in 1959, that these moths were immigrants from the south (Cai, 1990). The validity of this hypothesis was soon established and it was shown that the migrations were achieved by transport on the wind. In this chapter, we first review the earlier work on M. separata migration, and then summarise the results of a more recent research programme in which the migration has been observed directly with entomological radar.

Previous research

The migratory status of M. separata was demonstrated by a variety of techniques.

Helicoverpa spp. in Australia

Of the three species of Helicoverpa in Australia, two, the endemic H. punctigera and the cosmopolitan H. armigera, are major pests. In eastern Australia the two species frequently occur together, and can cause severe damage in the cropping regions of the southeast of the continent (Fig. 8.1). Summer crops affected include cotton, sunflowers, sorghum, soybeans, maize, and many vegetables and horticultural crops. Winter/spring crops include chickpeas, field peas and faba beans. It is often difficult to explain changes in the numbers of Helicoverpa spp. in the cropping areas, and this has led to speculation that immigration from non-cropping areas further inland could account for some of the discrepancies (Zalucki et al., 1986).

The pest status of Helicoverpa spp. arises from four characteristics exhibited to varying degrees by both species: high mobility, polyphagy, high fecundity and facultative diapause (Fitt, 1989). The evidence for migration in Helicoverpa spp. has been reviewed by Farrow & Daly (1987). They rated H. punctigera as the most migratory and H. armigera as the least, with the two major American pest species H. zea and Heliothis virescens intermediate.

There is compelling evidence that H. punctigera is highly migratory, and is probably an obligate migrant. Specimens of presumed Australian origin have been collected in New Zealand (Fox, 1978) and Norfolk Island (Holloway, 1977), indicating migrations of at least 2200 and 1600 km, respectively.

Introduction

To recognise and understand the environmental processes that lead to outbreaks of migrant pests, many environmental variables may need to be combined in a spatial context. Computer-based geographic information systems (GIS) make it possible to manipulate large spatially referenced data sets, including remotely sensed data on, for example, vegetation, rainfall and surface temperature. In this chapter, the extent to which these techniques have been applied to the ecology of migrant insect pests is discussed and ways in which they may be developed further are considered. The intention is not to provide technical information on the use of GIS and remote sensing, for which the reader is referred to standard texts (e.g. Curran, 1985; Burrough, 1986).

The use of GIS and remote sensing in entomology

GIS is the name given to the general techniques, and to specific computer packages, that can be used to register, store, manipulate and retrieve spatial data. The function of GIS is to integrate different types of spatial data. Spatial data are commonly held in computers in either vector or raster format and the relative advantages and disadvantages of each format are discussed by Burrough (1986). Vector data are a series of spatial coordinates defining points, lines and polygons that may be generated when maps are digitised, as well as specific point measurements, e.g. from meteorological stations, which are entered as tabular data. Raster data comprise a regular grid of cells, each having a particular value for a given variable. Satellite data are recorded in this format.

Introduction

The Cotton Bollworm Helicoverpa armigera(Hübner) is an agricultural pest occurring throughout Africa, southern Europe, the Middle East, Asia and Australasia (Commonwealth Institute of Entomology, 1968). An important factor contributing to its pest status is its ability to undertake long-distance migratory flights (review by Farrow & Daly, 1987; see also Chapter 8, this volume), that may occasionally exceed 2000 km (Bowden & Johnson, 1976).

Radar studies and field observations have shown that noctuid migration typically involves a characteristic sequence of behaviours (see e.g. Rose et al., 1985). Takeoff occurs shortly after sunset, with moths climbing at a rate of ~0.5–2.0 m s−1 until they reach the top of the temperature inversion where windspeeds are usually at a maximum, often 20–50 km h−1 (Drake & Farrow, 1988). The insects then maintain their altitude, continuing their wind-assisted progress until they descend at some time during the night or around dawn (Drake, 1985).

Because of the long distances migratory insects travel and their relatively small size, the study of migration at the ecological level presents major logistic and technological challenges. No less problematic, however, is the laboratory-based approach, in which the behavioural, physiological and genetic mechanisms that regulate migratory behaviour are investigated.

This chapter describes the contribution of laboratory studies to our understanding of the migratory behaviour of H. armigera.

Flight activity and the pre-reproductive period

Two components of migratory potential that are amenable to examination in the laboratory are flight activity and the pre-reproductive period (PRP).

Introduction

Rice is the staple food crop in China, which is the biggest rice producer and consumer in the world. In 1990, 33 million hectares of rice produced approximately 180 million tonnes of grain, representing 45% of the country's total grain output and 37% of world rice output. However, preharvest losses caused by insect pests are estimated at several million tonnes each year and a substantial proportion of these losses is attributable to the Brown Planthopper Nilaparvata lugens.

N. lugens has become an increasingly serious problem since the 1970s. Outbreaks have increased in frequency and the area regularly infested has extended into the Jianghuai region (between the Yangtze and Huaihe Rivers) and north of the Huaihe River. On average, some 13.3 million hectares of the crop are likely to be affected, with an annual loss of some half a million tonnes of grain (Table 17.1). In 1991, the worst year on record, severe damage extended over the whole rice-growing region and 17.3 million hectares of paddy fields were infested by N. lugens. In spite of great efforts at prevention and control, which were estimated to have saved approximately 9.8 million tonnes of grain, more than 2 million tonnes were still lost to these infestations. These efforts represented a substantial investment of manpower and material resources and the total loss sustained in that year was estimated at 2000 million Yuan (about US$ 400 million).

Introduction

Although there are many records in the literature of individual insect migrations within Africa and Europe (see e.g. Williams, 1958; Johnson, 1969; Cayrol, 1972; Rainey, 1976; Pedgley, 1982; Dingle, 1985), there have been few attempts to review seasonal redistribution on a continental scale (but see Farrow (1990) for the migration of Acridoidea). Here we summarise the (now very extensive) available information on long-range, windborne movements by winged insects in the various climatic zones of Africa and Europe, and show how these migrations are related to the weather and to seasonal changes of climate and prevailing wind.

Some insect species migrate only short distances (a few hundred metres to a few kilometres) between habitat patches (see e.g. Solbreck, 1985), but we are concerned here with migration on scales covering countries and continents. These distances are so great, and the flying speeds of most insects are so slow (usually less than 3 m s−1), that most migrations are dependent on the help of an external energy source: the wind. Persistent flight in a wind of, for example, 10 m s−1 can easily result in migrations of 300-400 km in one day or night. In many (but by no means all) species, these migrations are made during the few days before the onset of breeding (Johnson, 1969; see also Chapter 10, this volume), and the movements are often dominated by a single weather system (Pedgley, 1982; Drake & Farrow, 1988).