Abstract. A case study is presented of the autumn migration of the brown planthopper, Nilaparvata lugens (Stål), in the area of Nanjing in the People's Republic of China. The study was made using a high frequency (8 mm wavelength) radar and a net suspended from a kytoon. The observations confirmed that long‐distance return migrations occur in China in mid and late September, with N.lugens being carried on the prevailing north‐easterly wind towards the autumn infestation and overwintering areas of the species. After mass take‐off in the late afternoon or at dusk, the migrants flew for several hours during the evening, often in a dense layer which formed at heights between about 400 and 1000m above ground. These layers often had well‐defined ceilings corresponding to an air temperature of about 16°C. The migration height was above the top of the surface temperature inversion, i.e. the migrants did not fly at the height of the warmest air. The dense layer concentrations overflying the radar were backtracked to source areas up to 240 km away in the north‐east of Jiangsu Province. Planthoppers observed emigrating from the Nanjing area would reach areas in south Anhui Province or north Jiangxi Province if they flew for 12 h. There was a second period of mass take‐off at dawn. Insect layers sometimes formed but did not last longer than 1–2h. The present results were strikingly different from those previously observed in the dry season in the Philippines, where migratory flight durations were largely confined to periods of about 30min at dusk and dawn. Our observations are discussed in relation to the equator‐wards return migrations undertaken in autumn by other insect species, and the importance of these migrations for the maintenance of long‐flying genotypes in the overwintering populations is considered.

BACKGROUND: The fall armyworm (FAW), an invasive pest from the Americas, is rapidly spreading through the Old World, and has recently invaded the Indochinese Peninsula and southern China. In the Americas, FAW migrates from winter-breeding areas in the south into summer-breeding areas throughout North America where it is a major pest of corn. Asian populations are also likely to evolve migrations into the corn-producing regions of eastern China, where they will pose a serious threat to food security. RESULTS: To evaluate the invasion risk in easternChina, the rate of expansion and future migratory range was modelled by a trajectory simulation approach, combined with flight behavior and meteorological data. Our results predict that FAW will migrate from its new year-round breeding regions into the two main corn-producing regions of eastern China (Huang-Huai-Hai Summer Corn and Northeast Spring Corn Regions), via two pathways. The western pathway originates in Myanmar and Yunnan, and FAW will take four migration steps (i.e. four generations) to reach the Huang-Huai-Hai Region by July. Migration along the eastern pathway from Indochina and southern China progresses faster, with FAW reaching the Huang-Huai-Hai Region in three steps by June and reaching the Northeast Spring Region in July. CONCLUSION:Our results indicate that there is a high risk that FAW will invade the major corn-producing areas of eastern China via two migration pathways, and cause significant impacts to agricultural productivity. Information on migration pathways and timings can be used to inform integrated pest management strategies for this emerging pest.

The autumn migration of the rice leaf roller moth, Cnaphalocrocis medinalis Guenée, in eastern China was studied at two sites (one in southern Jiangsu Province and one in northern Jiangxi Province), using radar and aerial netting. It was confirmed that C. medinalis is a high-altitude nocturnal windborne migrant, with large numbers of moths taking-off at dusk and flying continuously for several hours. Migration was post-teneral and the females had immature ovaries. Maximum densities of the moths typically occurred between 250 and 550 m above ground, and layering was intense on some nights. Moth layers often occurred at an altitude where there was a wind-speed maximum. In early September, migrant C. medinalis from southern Jiangsu Province were carried on the winds in an approximately westward direction. However, the winds prevailing in late September and October took migrants from both sites towards the south-west or south. Forward trajectories for C. medinalis and other insects emigrating from northern Jiangxi during October indicated that they were able to reach the tropical rice-growing areas near the south China coast, where year-round breeding would be possible. Incidental observations on the migratory flight of other lepidopteran pests, including Mythimna separata (Walker), Spodoptera litura (Fabricius), Ctenoplusia agnata (Staudinger) and Agrotis ipsilon (Hufnagel) (all Nocturidae), Spoladea recurvalis (Fabricius) and particularly Omiodes indicata (Fabricius) (both Pyralidae) are presented.

Rice planthoppers and associated virus diseases have become the most important pests threatening food security in China and other Asian countries, incurring costs of hundreds of millions of US dollars annually in rice losses, and in expensive, environmentally harmful, and often futile control efforts. The most economically damaging species, the brown planthopper, Nilaparvata lugens (Hemiptera: Delphacidae), cannot overwinter in temperate East Asia, and infestations there are initiated by several waves of windborne spring or summer migrants originating from tropical areas in Indochina. The interaction of these waves of migrants and synoptic weather patterns, driven by the semi-permanent western Pacific subtropical high-pressure (WPSH) system, is of critical importance in forecasting the timing and intensity of immigration events and determining the seriousness of subsequent planthopper build-up in the rice crop. We analysed a 26-year data set from a standardised light trap network in Southern China, showing that planthopper aerial transport and concentration processes are associated with the characteristics (strength and position) of the WPSH in the year concerned. Then, using N. lugens abundance in source areas and indices of WPSH intensity or related sea surface temperature anomalies, we developed a model to predict planthopper numbers immigrating into the key rice-growing area of the Lower Yangtze Valley. We also demonstrate that these WPSH-related climatic indices combined with early-season planthopper catches can be used to forecast, several months in advance, the severity of that season's N. lugens infestations (the correlation between model predictions and outcomes was 0.59), thus allowing time for effective control measures to be implemented.Keywords Nilaparvata lugens · Windborne insect migration · Atmospheric circulation · Rice pests · Planthopper risk prediction · Western Pacific subtropical high-pressure system Key message• East Asia has experienced a resurgence of serious rice planthopper outbreaks in recent years. In one of the worst-affected areas, the Lower Yangtze Valley of China, the number and timing of brown planthopper immigrants from further south has a major influence on the risk of local outbreaks.• Here we show how seasonal outbreak risk can be predicted from indices of the intensity of the western Pacific subtropical high-pressure system, a major atmospheric circulation system that drives the synoptic weather patterns affecting planthopper immigration.Communicated by V. Gagic. Electronic supplementary materialThe online version of this article (https ://doi