Tracking locusts – in Yemen and far beyond


George Popov MBE spent his career in locust research and control and despite retirement in 1984, has remained a leading international expert in this field. From 1958-64 he served as team leader of a UN-sponsored survey of desert locust breeding grounds from Senegal to Bangladesh and Turkey to Tanzania. He travelled extensively in Yemen in 1962. This article was published in the British Yemen Society's journal, 1998.

Man’s encounter with locusts is of very old standing and probably from the earliest times they were an item on his menu — a dietary practice which still survives in parts of the world, including Yemen, to this day.

With development and the spread of agriculture and animal husbandry, the importance of locusts greatly increased. Locusts and their close relations, grasshoppers, are herbivorous with a lively appetite, daily consuming a quantity of food equivalent to their own weight; indeed at times their appetite becomes positively voracious and they can devour many tons of vegetation, cultivated and wild, especially pasture, to create a very serious problem.

Initially, Man had no defence against locusts which were therefore seen as a sign of God’s displeasure and wrath. Witness, for example, the plague mentioned in the Old Testament as having afflicted Egypt — in response to Moses’s prayers for help against the Egyptians at the time of the Jewish Exodus — resulting in a massive loss of crops and famine.

Gradually, however, Man came to regard locusts not merely as the agents of divine wrath and punishment but as a natural phenomenon which he could confront and control to defend his means of subsistence. Nevertheless, for a long time this defence was of a purely mechanical nature — the digging of trenches to trap and bury flightless bands of locusts or the torching of both them and winged swarms when settled densely on bushes. It is only from the beginning of this century that strategies of locust control assumed greater sophistication and methods of monitoring locust populations and of conducting chemical and biological control were introduced.

There are several thousand species of locusts and grasshoppers. They are a world-wide phenomenon; only the cold Arctic and Antarctic regions are devoid of them. The Locust and Grasshopper Agricultural Manual (COPR, 1982), which lists the agriculturally more important species, cites 30 as occurring in the north of Yemen and 23 in the south but allowing for the less common species the total is probably nearer one hundred. Three of these species are classified as locusts: the desert locust (Schistocerca gregaria), the migratory locust (Locusta migratoria) and the tree locust(Anacridium melanorhodon arabafrum). The distinction between locusts and grasshoppers was first studied and described by the late Sir Boris Uvarov, an eminent acridologist and entomologist who set up the Anti-Locust Research Centre in London and was my mentor in the 1940s. The distinction hinges on the capacity of locusts when densely crowded to undergo behavioural and physiological changes which induce them to form cohesive marching bands of wingless ‘hoppers’ in their immature stages, and cohesive swarms as winged adults; both stages are mobile but, of course, it is the winged adult swarms which migrate long distances. While grasshoppers are sometimes very numerous and can cause considerable damage to crops and pastures, locusts by virtue of their high numbers, density and mobility constitute the main danger and require the greatest effort to contain them.

Considerable progress has been made in developing effective control strategies against most species of locusts and grasshoppers, the fundamental approach being discovery and identification of areas where survival, concentration and breeding can occur. In the case of locusts this can result in band and swarm formation: i.e. the transformation from low density individuals of the so-called ‘phase solitaria’, to dense populations living in dense groups of the ‘phase gregaria’. For most species such source/outbreak areas were identified and preventive measures ensured by programmed survey and contact operations — except in the case of the Desert Locust whose source and outbreak areas remained largely unknown until the end of the 1950s. On the principle of ‘know thy enemy’ Sir Boris Uvarov recommended a widespread and intensive ecological survey of the total area of the distribution of the Desert Locust, extending in a broad belt from the Atlantic to central India, by a team including a locust expert and a botanist. The writer and a Swiss botanist, Charles Rossetti (later replaced by W Zeller) were recruited for this project, initially under the auspices of FAO and UNESCO but later under UNDP Over a seven year period 1958-1965 the team sought to visit all the more important known or suspected locust habitations. In most areas the team was accompanied by a local liaison officer with some experience of locust activity in his country. In Yemen which we visited in 1962 one such officer was Nasir al-Mu’afa who, following retirement, continued to work with the Yemeni Ministry of Agriculture as senior consultant until quite recently.

Our survey made some important discoveries about the biology and ecology of the Desert Locust which in due course helped to develop an effective strategy for the monitoring and control of locust populations. In its hostile, arid habitat the Desert Locust survives by virtue of its high mobility (day-flying gregarious swarms and night-flying solitarious individuals) and also by its ability to recognise habitats suitable for its survival, multiplication and, occasionally, gregarisation. This ability is particularly important for solitarious phase locusts (pictured, right) which inhabit the central, more arid half of the distribution area where habitats are small, highly localised and seasonal, coming into existence as a result of sporadic rainfall. On the whole, these habitats constitute a small fraction of the total area so that when they dry out, locusts need to find alternative habitats in order to survive. We are still uncertain how they do this so effectively, but the fact remains that a succession of suitable habitats enables locusts to increase their numbers and density, to convert into mobile bands and swarms and to spread over larger areas — at worst leading to plague conditions. Against this background and given the difficulty of locating low density locust populations, the optimal strategy for tracking locusts is first to monitor their habitats as they come into existence in response to the vagaries of rainfall, and to examine those that have the best locust potential. Preventive control is directed against populations in the course of gregarisation or which are already gregarious. During the thirty-five years since our survey, the distribution and potential of locust habitats has been investigated to a considerable extent and the details elaborated on maps and atlases, biomodels and computerised data bases for use by the FAO’s locust information bureau and by regional and national locust control organisations. Nevertheless, due to financial and administrative constraints and the sheer difficulty and complexity of the problems, the early stages of a plague upsurge are still occasionally missed.

In Yemen the Desert Locust problem is of considerable importance and diversity. reflecting the wide range of geographical and ecological conditions in the country. Yemen lies at the cross-roads of swarm migrations arising in source areas in the Eastern Region (i.e. India, Pakistan, Afghanistan, Iran and Oman) and the Central Region (the rest of Arabia, the Near East Sudan and the Horn of Africa). Depending on the locust situation, prevailing winds and the rains, Yemen may receive swarms from any of these sources and in its own turn contribute to the volume and number of the migrant swarms. During the twenty-five year period 1939-63 many countries in the region witnessed some swarm activity in most years and the bulk of locust populations were in the gregarious, swarming phase; swarms had visited almost all parts of Yemen, including the highest parts of the central massif and the interior of the Rub’ al-Khali desert. However, the coastal plain of the Tihama and the southern coast witnessed the greatest level of activity and the concentration of bands and swarms, some of which were produced by local breeding while others arrived as migrants from neighbouring countries. Months which saw the highest incidence of breeding were July to November, corresponding to the period of the summer monsoon rains and sometimes extending into early winter on the coast. In the hinterland breeding was less common and largely confined to the sandy beds of wadis.

During years of recession, for example the period 1964-1986 (excluding 1968 and 1978 which were years of swarming activity), locusts in Yemen were very largely confined to the coast, being uncommon in the interior and totally absent in the cool highlands.

Incidence of breeding closely reflected periods of rainfall, highlighting the dependence of breeding on rain. The natural habitats where locust breeding occurs are the pastures (khabt), fallow land and fields of millet and sorghum (dukhn and dhurra) with an abundance of annual herbs and weeds such as Heliotropium, Dipterygium glaucum, Chrozophora, Tribulus and Indigophera, growing on light sandy soil where locusts and ‘hoppers’ tend to breed, aggregate and ultimately to gregarise.This, for example, occurred inland in the area between Mareb and Harib during the winter-spring of 1950, giving rise to uncontrolled primary locust swarms; later these swarms migrated to Somalia, bred on a scale beyond the resources of the locust control organisation and gave rise to a plague which lasted for the next twelve years until the early 1 960s and affected much of eastern Africa and the Arabian peninsula before spreading to neighbouring countries.

Compared with the Desert Locust, the other two — the migratory and tree locusts are of relatively minor importance: they do not form persistent migrant swarms and even in the absence of control, disperse and die down due to natural mortality. They are also ecologically quite distinct. The migratory locust — a subspecies close to the African Locusta migratoria migratorioides is a strict grass feeder that occurs in the grasslands of some of the major wadis of the Tihama; occasionally when very numerous, the denser populations may gregarise to form bands and swarms which will attack and harm cereal crops of millet and sorghum to the vocal alarm of their owners! However, in the light of my own experience, such infestations decline naturally, although some local crop damage may occur. The tree locust —Anacridium melanorhodon arabafrum — is an arbusticulous shrub dweller whose natural habitat is in the thickets and shrublands of Acacia (samr and talh) and the ‘ilb trees (Ziziphus spinachristi) found here and there in the foothills of the Tihama. When very numerous, tree locusts can defoliate their natural habitats and may thus need to seek sustenance elsewhere. In the process they may aggregate and form bands and swarms which may then invade crops (e.g. fields of millet, orchards of citrus, apricot and pomegranate, and vineyards) and can cause considerable damage. This is usually a local phenomenon but farmers are likely to need expert help to contain it.

Despite major technical advances in locust control during the past thirty-six years (1962-1998) the situation in Yemen and elsewhere in the region appears to have changed very little. This is surely a measure of the complexity of the problem and the difficulty of organising an effective, long-term solution. But my final word must be to express deep gratitude for the privilege — granted during the rule of Imam Ahmad bin Yahya and, in their turn, by the British authorities in Aden — of visiting this fascinating, now united country and of experiencing the friendliness and hospitality of its warm-hearted people in 1962 and also during subsequent visits in 1972 and 1985. It is a memory which I shall cherish always.

November 1998