Forecasting of Rodent Outbreaks: Models and the Real World1

Theoretical and practical purposes of mathematical modelling are discussed. Of these, special attention is given to large scale simulation models and regression models. It is pointed out that, for proper use of simulation models for predictive purposes, it is necessary to realize the uncertainties associated with such model predictions. These uncertainties are mainly due to insufficient knowledge of parameter values included in the simulation model. As these model predictions are made iteratively (i.e. by short time steps), biases in parameter estimates have, with increasing length of simulated time, an accumulating effect on the biases in the output variables. Regression models are described and methods for finding the best mathematical model are discussed. Assumptions regarding statistical properties of the data from which the regression model is constructed are discussed. It is pointe out that valid predictions are made only within the domain defined by the data Examples of regression models for forecasting rodent outbreaks are discussed. Some further developments are proposed. Results from a model emphasizing spatio-temporal heterogeneity in a heterogeneous region are discussed. Based on this model conclusions regarding what type of landscapes are likely to cause rodent outbreaks are deduced. Furthermore, proposed control treatments are suggested by analyzing the model. It is concluded that larger regions should be composed of suitable and non-suitable rodent habitats interminded with each other. It is suggested that treatments for increasing the local extinction rate be applied uniformly over the whole region, but varyjng in time as much as possible and not correlated with the density in local habitats nor in the entire region. Within these limits, treatments should be applied during periods of low mobility and low reproduction.     

Linear Regression Models for the Effect of Climatic Factors on the Level of Attack by Agrotis sogetum1

Based on historical records, cutworm attack levels for the years 1906 to 1979 have been estimated on a scale from 1 to 11 for Denmark as a whole. By means of linear regression methods the attack levels were related to monthly values of ordinary climatic variables. A resulting model of 8 variables could account for 64% of the total variation of attack level. The model showed a reducing effect on attack level of the amounts of precipitation during late winter and during May. For June, July and August the reducing effect of the numbers of precipitation days was stronger than that of the amounts of precipitation. Soil moisture as a mortality factor for larvae and pupae is suggested as the reason for the influence of precipitation during late winter and May, while rain in the summer may decrease flight, mating and oviposition of the adults.     

Aerial Transport of Pests and Pest Outbreaks1

Pests of arable crops tend to show a high degree of migratory activity. This is particularly so in northern Sudan where the variable rainfall produces habitats which are short-lived and unpredictable in their frequency and location. Insects whose flight activity is adapted to enable them to move on the same wind-system which produces the rains are frequently pests of the crops of these climates. Outbreaks of such pests are usually a function of the numbers which colonise the crops, and this in turn depends not only on the success of breeding in the source areas but also on increases in density which occur whilst they are airborne. The importance of the increases in airborne densities which occur within the inter-tropical discontinuity and the cold-air outflows ahead of convectional storms to pest outbreaks in the Sudan Gezira is described.     

Role of Population Quality in Forecasting for Plant Protection1

In order to assess the future abundance of a pest as the basis of the damage to be expected, actual numbers of individuals will not suffice. It is not only pre-damage mortality of parasitized or diseased individuals that modifies threshold levels or the results of forecasting future losses. In addition, the quality of populations must also be considered, as it affects the expression of behaviour, reproduction and survival. This is illustrated by using two parameters: the effect of enzootic diseases and differences in yolk content of lepidopterous eggs, both of which have a strong influence on the validity of prognoses. The concluding recommendation is to find and to utilize qualitative differences between individuals as an aid to developing an improved method of forecasting insect development. Such progressive pest management would include forecasting the development of both pest and beneficial organisms and it would optimize means of proper manipulation. Le dénombrement d'un ravageur ne suffit pas à la prévision de l'évolution future des populations de celui-ci. A part la mortalité des insectes parasités ou malades, qui entraine la modification des seuils de tolérance et des prévisions de pertes, il faut tenir compte de L'effet de la qualityé des populations, dans la mesure ou celle-ci influe sur le comportement, la reproduction et la survie des ravageurs. Deux paramétres exercant une influence importante de ce genre servent dillustration; il s'agit de L'effet des maladies enzootiques et des differences de teneur en vitellus des ceufs de lépidopteres. Il serait donc utile de recenser les differences qualitatives entre individus, ce qui pourrait servir à L'amélioration des previsions sur le developpement des insectes. Ceci est realisable aussi bien pour les insectes nuisibles que pour les auxiliaires et permet d'assurer une approche dans le sens du management.     

Meteorology and Desert Locust Forecasting1

The now very considerable understanding of weather effects on breeding and movement by the desert locust has been used to prepare a forecasting manual. It is a two-volume, 400-page publication that describes, explains and illustrates the principles of modern forecasting, and includes not only much material on the biology and ecology of this species but also many new biogeographical studies of past locust developments that involved a variety of synoptic systems in some 60 countries of northern and eastern Africa and south-western Asia. It can be useful to all who are concerned with the effects of weather on insect movement.     

Forecasting and Warning in Danish Agriculture1

In Denmark, the principle is maintained of forecasting and warning against pests and diseases which are not easily recognized by the farmer himself. The forecasting and warning system is based upon close cooperation between the farmers' unions, a well-organized network of plant husbandry advisers and the State Plant Pathology Institute in Lyngby. Most field data are collected by the advisers and forwarded to Lyngby from where announcements on current attacks are distributed as appropriate. Realizing the considerable local variations in the occurrence and course of the attacks, plant pathologists mainly notify the advisers, who can then intervene, if necessary, with a local warning for their own districts. Potato blight, however, is an exception, country-wide forecasting being broadcast every year. This may also be the case with warnings againts aphids. Potato blight forecasting has been in operation since 1935 and warnings against aphids in beets (causing virus yellows) since 1957. Each year about 10 different pests and diseases are warned against. Au Danemark, le principe prévaut d'assurer les prévisions et les avertissements pour les déprédateurs qui ne sont pas facilement dépistés par I'exploitant. Le système se base sur une coopération étroite entre les syndicats dexploitants, un réseau bien organisé de conseillers agricoles, et 1'Institut national de phytopathologie à Lyngby. Les observations de plein champ sont realisees par les conseillers qui transmettent les resultats a Lyngby. L'Institut assure alors la dissemination des informations sur les attaques en cours. Compte tenu de la forte variation regionale de L'incidence et de I'evolution des attaques, les phytopathologues avertissent les conseillers qui decident s'il y a lieu d'emettre des avertissements dans leur district. Le mildiou de la pomme de terre, cependant, est considere au niveau national et, suivant les cas, egalement les pucerons. Les previsions contre le mildiou sont en place depuis 1935, et contre les pucerons de la betterave (vecteur du yellows virus) depuis 1957. Chaque années, une dizaine de ravageurs et maladies font L'objet d'avertissements.     

FAO's Need for Rodent Ecology,Population Dynamics and Forecasting Data1

During the past decade, technical assistance in rodent damage control has tended to utilize less and less resources per problem, while the results expected per unit of resources used has rapidly mounted. The developing world, with its multitude of needs, lacks personnel, funds and organizational infrastructure for self-help. The donor world, with its inflation, requires justification and the prospect of probable achievement of objectives prior to sponsoring assistance. On the basis that these trends are likely to continue, it is proposed that: 1) national rodent damage control assistance be action-oriented; 2) multi-country assistance rapidly evaluate damage control strategies of stubborn, common problems; 3) both national and multi-country assistance projects be designed with the flexibility to accept unsatisfactory results in objective achievement and to move to a slower, more certain route, such as development of ecological and population dynamics data; 4) co-operation among agencies, donors and developing countries be increased for more effective use of scarce resources; and, 5) an ad hoc advisory group of vertebrate pest control specialists meet in 1977 to give guidance in further strategy refinement for development assistance. With the hope of avoiding frequent use of the flexibility of assistance project design, data from the more developed world on ecology, population dynamics and forecasting should be examined in all developing country projects as to possible useful extrapolations which will increase the likelihood of achieving objectives within the available resources.     

Forecasting the Spread of Migrant Insect Pests1

When there is reasonable evidence that an insect pest is a migrant, the forecasting of outbreaks and the planning of control measures need an information service that can describe the continually changing distribution of the pest. As records are gathered the likely changes in a particular region and season become clear and forecasting by analogues is possible. More precise forecasts can be made when the mechanism of migration is understood, such as downwind movement. Effective forecasting of migrant pests needs continuity in the assessment of redistribution of all major populations of the species over a wide area - preferably the whole distribution area. Experience with forecasting infestations of certain tropical migrants, locusts and armyworms, suggests methods of organising and interpreting information from survey networks, so that the effects of migration can be taken into account when preparing warnings. Pour les insectes migrateurs, il est important d'établir un service d'information appeléà prévoir la distribution constamment changeante du ravageur, les gradations, et partant, les mesures de lutte. L'accumulation des données pour les regions et saisons differentes permet d'etablir des prévisions plus precises, en utilisant des modeles. Une meilleure comprehénsion du mécanisme de migration, découlant par exemple de I'étude de I'influence du vent, permettra également d'affner les prévisions. Pour émettre des prévisions cables, il faut assurer la continuityé des observations sur tous les deplacements majeurs des populations de L'espece, en tenant compte de grandes etendues embrassant si possible L'ensemble de L'aire de dissémination. Les expériences acquises par l'étude de certains migrateurs tropicaux (criquets et noctuelles) servent a mieux definir les méthodes dorganisation et d'interprétation des données accumulees par les reseaux de surveillance. Elles permettent d'émettre des avertissements qui tiennent compte des phénomenes de migration.     

Crop Loss and Disease Incidence Classified by Climatic Parameters1

A method of analysing survey data, collected to obtain information on various aspects of cereal crop performance, especially disease incidence, plant population, and yield, has been developed to allow estimates of potential yield, and hence total loss, to be made. The method also allows the losses to be partitioned and the components attributed to different stages of the crop's growth - establishment, tillering and grain formation. The farms used in the survey have been classified according to winter, spring and summer temperatures and rainfall. Attempts have been made to associate trends in disease incidence and losses occurring at different times of the year with the climatic parameters.     

A Synoptic Approach to Regional Forecasting in Plant Protection (The Hungarian Forecasting System)1

Forecasting is of primary importance for the effectiveness and profitability of plant protection on a regional and on a farm level. Farmers should be guided by regional forecasts outlining major trends of pest infestations and epidemics. However, additional farm observations are also needed to adapt general trends to local farm conditions. The Hungarian National Plant Protection Organization embraces the Agrochemistry Centre and 20 regional Stations. Plant protection forecasting activities are organized and supervised by the Forecasting Department of the Centre and every Station has a Forecasting Group. Data acquisition is the first step in the forecasting procedure. It is organized on an ecological rather than on an administrative basis. A national system of ecological forecasting regions has been set up and the results of observations are processed on a regional basis. There are permanent and temporary sampling localities and additional phenological stations within the regions. At the permanent sampling sites and at the phenological stations, observations are carried out according to a national sampling plan consisting of detailed routine instructions for quantitative and qualitative surveys. Three types of biological information are included: appearance of pests/diseases and their seasonal development, density distributions of pests/diseases over forecasting regions in selected periods of the year, and crop plant phenology. Information is coded at the Stations and forwarded to the Centre by telex every day during the growing season. At the Centre, the data are plotted immediately into maps combining the ecological forecasting regions and a national grid system of the country. Final data processing is done on a daily basis by using a synoptic mapping technique and forecasts are regularly released by means of several media to the large farms, home gardeners and private growers. The media include forecast circulars, at irregular time intervals, released by the Centre and the Stations, as well as weekly journals of agriculture or horticulture and national and local daily newspapers. The Hungarian Radio and Television also broadcast plant protection forecasts weekly and fortnightly, respectively. La prévision est trés importante puisqu'elle détermine L'efficacité et la rentabilité d'une mesure de protection. Les exploitants devraient pouvoir bénéficier d'un service de previsions régionales faisant état des grandes lignes de I'évolution des infestations et des épidemies. Ces données doivént cependant être doublees d'observations faites au niveau de I'exploitation, afin dajuster I'information generale aux conditions locales. En Hongrie, I'organisation nationale phytosanitaire comprend le Centre de la protection des végétaux et d'agrochimie, ainsi que 20 stations régionales. La prévision est assuree par le département du Centre qui est rélie aux groupes de specialistes pour la prévision, installés dans chacune des stations. L'acquisition des donneés, qui constitue la premiere étape de la prevision, repose sur un concept écologique. Le pays est ainsi subdivisé en régions écologiques et les observations sont traitees au niveau régional. Dans chaque region, il y a des stations d'echantillonnage permanentes ou temporaires, ainsi qu'un certain nombre de stations phenologiques. Les observations aux stations permanentes et phenologiques se font conformément à un plan national, qui precise la fqon de proceder aux inspections quantitatives et qualitatives de routine. Les données nécessaires comprennent: 1) I'apparition des déprédateurs et leur evolution au cours de la saison; 2) la distribution et la densité des depredateurs dans les regions a certaines epoques precisees; 3) la phénologie des plantes. Elles sont enregistrees aux stations et sont quotidiennement transmises par telex au Centre pendant la periode de vegétation. Au Centre, les données sont rapportées sur des cartes representant les régions écologiques et qui sont superposees sur une carte du pays. Le traitement final des données est assuré tous les jours par le recours à une technique synoptique de cartographie. Les prévisions sont communiquees selon les nécessites aux grandes exploitations et aux exploitants privés par bulletins expédies par le Centre et les stations, ou par la voie de revues hebdomadaires dagriculture ou dhorticulture et de journaux nationaux ou locaux. La radio et la television transmettent des prévisions phytosanitaires à intervalles d'une ou deux semaines.     

Forecasting the Incidence of Aphids Using Weather Data1

Aphids have been trapped continuously at several heights at Aberystwyth (United Kingdom) since 1969. Relationships have been found between prior meteorological parameters (normally recorded at Agricultural Meteorological Stations) and numbers of aphids flying during both summer and autumn peaks, and with the first occurrence in the trap each year. Numbers of cereal aphids caught in the autumn are associated with previous winter and spring temperatures and particularly with summer rainfall. Cereal aphid numbers are also associated with the yields of their summer grass hosts which in turn are related to summer rainfall, thus providing a causal relationship. The value and accuracy of using meteorological parameters to forecast the incidence of aphids of agricultural importance is discussed.     

Disease Intelligence and its Role In Disease Forecasting1

Disease intelligence information is received at Bristol by telex from plant pathologists in all regions of England and Wales. Weather information recorded at synoptic stations throughout the country is fed into the Meteorological Office computer at Bracknell, and from given parameters daily forecasts are issued for a number of diseases including apple scab, potato blight, barley mildew, barley leaf blotch, wheat septoria and wheat eyespot. All this information is collated, printed and distributed widely to plant pathologists in the advisory services, research establishments and various sections of commercial manufacturing and sales organisations. The report is broadcast on BBC radio, featured on teletext (BBC Ceefax), viewdata (Prestel) and Telephone Information Services (TIS), and published in a number of national and regional agricultural journals. Disease intelligence information aids in the interpretation of disease forecasts and equally disease forecasts aid in advising on the need for pesticide application.     

Monitoring and Forecasting Methods for Codling Moth Management in the United States and Canada1

With the renewed emphasis on integrated control in fruit growing, new monitoring and forecasting methods had to be developed to achieve efficient and economical but less disruptive codling moth control. Attractant traps for monitoring adults have long been in use; pheromone traps are a recent addition. They allow for easy monitoring of moth populations but interpreting the significance of catches remains difficult. Pheromone trap catches have been correlated with the seasonal phenology and with estimates of fruit injury. In the United States and Canada critical catch levels below which no treatments are advised have been determined. In addition, new forecasting methods are being developed which include simple physiological-time tables, phenological models and « strategic » population models. Computer-based information delivery systems allow access to these models from the field. It is too early to assess the total impact of these new monitoring and forecasting programs on control practices. Aux Etats-Unis, les systemes de lutte integree suscitent un inter& croissant. Dans les vergers, toute approche dans ce sens requiert un affinement des methodes de recensement des populations de carpocapse des pommes et des poires, de mbme qu'une amelioration des previsions. Depuis longtemps, plusieurs types de pieges attractifs ont ete utilises, mais les pieges a pheromones sont d'introduction recente. Ceux-ci permettent de reveler d'une facon remarquable I'activite des populations, alors que I'interpretation de la signification economique des captures demeure difficile. Aux Etats-Unis et au Canada, il a ete possible d'etablir une relation entre les captures realisees a I'aide de pheromones et les degsts observes sur fruits. Ceci a permis d'etablir des seuils critiques au-dessous desquels il n'est pas: recommande de recourir a une mesure de protection. Parallelement au developpement de tables de survie, de modeles phenologiques et stratkgiques, il a ete etabli un systeme d'information par ordinateurs. I1 est encore trop tat pour evaluer la valeur de ces nouveaux éléments.     

Forecasting Aphids in Field Crops (Swiss Model)1

A new prognostic system for aphids in field crops, based on long-term ecological investigations, was developed in Switzerland during the years 1971 - 1976 and tested successfully in practice in 1975 and 1976. Within the region, population densities of aphids and mortality caused by predators, parasitoids and fungus diseases were measured weekly at different places. Sampling was started in early spring (parthenogenetic overwintering of the aphids was never observed) in permanent and temporary grassland and continued in annual field crops after aphid migration. Typically, in years during which aphids caused damage (1973, 1975, 1976), there was a late progression of aphid population densities in permanent grassland in spring. But, during years without damage (1971, 1972, 1974), an early and intensive aphid development was observed in permanent pastures in spring. Thus, the intensity of the spring development of aphids in permanent grassland is of general importance for the development of the pests later on in annual field crops. This conclusion was considered in the forecasting system for which guidelines have been developed, enabling the Regional Plant Protection Services to establish exact prognoses. Farmers can then start making observations in their fields, in order to find the right moment for insecticide applications. Un nouveau systemè de prévision pour les pucerons des grandes cultures a été developpe en Suisse au cours des annees 1971–1976; il repose sur des recherches écologiques de longue durée et sa mise en place en 1975 et 1976 s'est révelée très utile. Les densités de populations de pucerons dans la région ainsi que la mortalité occasionée par les prédateurs, les parasitoides et les maladies ont été déterminées chaque semaine à des sites differents. L'échantillonnage commence au printemps (L'hivernation des populations parthénogenétiques n'a jamais été observée) dans les prairies permanentes et temporaires et se prolonge dans les cultures annuelles apres la migration des pucerons. Les annees au cours desquelles les pucerons ont provoqué des déghts importants (1973, 1975, 1976) sont caractérisées par un developpement tardif des populations de pucerons dans les prairies permanentes au printemps. A L'inverse, les annees caractérisées par I'absence de dommages (1971, 1972, 1974) ont vu un developpement printanier intensif des pucerons dans ces mêmes sites. Ainsi, L'intensité de ce développement exerce une incidence sur l'évolution ultérieure du ravageur dans les cultures annuelles. Cette conclusion a été incorporée dans les directives pour la prévision dont I'établissement a permis aux services regionaux de la protection des végétaux detablir un pronostic precis. Les agriculteurs peuvent ensuite observer leurs champs afin de determiner la date optimale pour le traitement chimique.     

Farm-level Forecasting: its Role and Organization in Plant Protection1

Forecasting in plant protection has two principal objectives. It should support decision making both on the necessity of control measures and on the correct timing of their application. Regional forecasts may guide plant protection activities on farms, but they have to be supplemented by local observations on the key pests of the most important crops. It is concluded that the organization of farm-level forecasting is closely related to certain characteristics – degree of harmfulness and oligophagy – of pests and diseases. Regular control is necessary of the most harmful pests whose economic thresholds lie below the general ecological equilibrium, and the correct timing of treatments should be accurately determined by means of local forecasting techniques. On the other hand, the harmful organisms whose economic thresholds lie above the general ecological equilibrium require comparison between the economic thresholds and the actual local densities. These observations should be carried out either by individual workers or by temporary working groups, depending on objective and on the area to be controlled. La prevision phytosanitaire doit satisfaire aux deux exigences principales suivantes: determiner la necessite d'une intervention et en fixer la frequence. Les previsions regionales fournissent des indications aux producteurs, mais ceux-ci doivent assurer leurs propres observations locales sur les depredateurs les plus importants. La facon de concevoir les previsions au niveau du producteur depend de I'impact economique et de L'oligophagie des organismes nuisibles concernes. Si les seuils economiques de ceux-ci se situent en general au-dessous d'un certain equilibre ecologique, il faudra prendre les mesures de protection et les systemes de previsions locales serviront a determiner les dates precises des interventions. En revanche, lorsque les organismes nuisibles ont des seuils de tolerance economique depassant generalement I'equilibre ecologique, il faudra comparer les seuils economiques a la densite des populations enregistree. Suivant I'ampleur de la tiche, celle-ci pourra être accomplie individuellement ou par un groupe de travail.     

Plant Protection Forecasting Service in the Federal Republic of Germany1

In the Federal Republic of Germany, plant protection is a matter for the several Federal Lander. Each of the regional plant protection offices has built up a warning system of its own, although there are no essential differences between them. The warnings are based on field investigations which are mostly carried out by technicians. However, it is not possible to inspect all fields, orchards, etc., as the areas of crops are too vast. Warnings follow a general pattern: information on the situation, followed by conclusions on further development; recommendations for control measures; and, most importantly, indications of possible undesired or even dangerous side-effects and how to avoid them. The warnings are published and also distributed by telephone and radio. Warnings on sugar beet in the Nordrhein area are distributed by the sugar factories to each of the farmers concerned. The forecasting service can certainly be improved by scientific progress. On the other hand, practical improvement is indispensable. The main problem is how to get more information from the field. There is little hope that the personnel situation will change fundamentally, but every effort must be made, in view of the importance of the forecasting service. En République féderale d'allemagne, les Lander sont chargés de la protection des végétaux. Les bureaux régionaux ont développe leurs proprés systemes de prevision, mais ceux-ci sont neanmoins très semblables. Les avertissements sont basés sur des observations de plein champ qui sont réalisées pour la plupart par des techniciens. I1 n'est pas possible d'inspecter tous les champs, vergers, etc., en raison des enormes surfaces en jeu. Les avertissements reposent sur un système normalisé comprenant I'information sur la situation présente, suivie d'une évaluation de I'evolution ulterieure, dune recommandation sur les methodes de lutte, d'une indication sur les effets secondaires indesirables ou dangereux, ainsi que des moyens de les eviter. Les avertissements sont expedies sous forme de bulletins, ou communiques par telephone ou par radio. Dans la région de Nordrhein, les avertissements concernant la betterave sont distribues aux exploitants par les sucreries. Le système de prevision peut certainement btre ameliore, surtout sur le plan pratique en beneficiant des progres scientifiques qui sont realises. I1 est essentiel de mutiplier les observations de plein champ, mais il est improbable que les services responsables puissent disposer du personnel necessaire. Cependant, face à I'importance de la prévision, il faut tirer le meilleur parti des moyens disponibles.     

The Forecasting System for Plant Protection in the USSR: Its Theory,Methods and Organization1

In the USSR, the system of forecasting the occurrence of crop pests and diseases consists of long-term, yearly or seasonal forecasts, and a warning service. They allow for prophylactic plant protection to be adequately planned and organized both on a country-wide and a regional scale. Theoretically, all kinds of forecasts are based on a knowledge of the population dynamics of pests. The criteria used to develop forecasts are the state of certain climatic and energetic factors during previous seasons, which determine the morphophysiological features of populations and the way they will react to the environment in the future. On this basis, mathematical formulae can be developed and the management of pest population dynamics can be planned. The main biological information required for forecasting is collected by the State Plant Protection Service. Data on climatic conditions and the state of crops are supplied by the Hydrometeorological Service. Methods of data collection, treatment and interpretation are developed by research institutes according to the accepted model of pest population dynamics. En URSS, les previsions se font a long terme, a echeance d'une annee et sous forme d'avertissements saisonniers. La protection des plantes peut, de ce fait, etre planifiee a I'echelle nationale et regionale. En principe, toutes les previsions sont basees sur la connaissance de la dynamique des populations de ravageurs et maladies. Les criteres utilises pour la preparation de previsions comprennent le recours a une serie de facteurs climatiques et energetiques qui ont prevalu durant les saisons precedentes. Ceux-ci determinent les caracteristiques morphophysiologiques des populations et la faGon dont elles vont repondre aux conditions de I'environnement. Ceci permet de developper des formules mathematiques de prevision et de planifier le management des populations des depredateurs. La plus grande partie des donnees necessaires a ces previsions sont recueillies par le Service de la protection des vegetaux. Les donnees sur le climat et sur la phenologie des cultures parviennent au Service hydrometeorologique, alors que les methodes de lutte sont developpees par les instituts de recherches, sur la base de modeles de la dynamique des populations.     

Search for resistance to phytophthora root rot in West Indian avocado rootstocks in the Canary Islands1

Avocado root rot caused by Phytophthora cinnamoni is a major limitation for avocado expansion in the Canary Islands, as it is in other producing areas of the world. To date, no germplasm with resistance potential has been found among West Indian avocados, which would convert this type into an ideal rootstock for use in infested areas such as the Canary Islands. Our aim was a thorough screening of the local West Indian population, descendants of seedlings brought to the Canaries from the Caribbean in the early 1800s. Our selection methods were based on those used by Zentmyer in California (US) since 1960 (nutritivesolution tank, with controlled aeration, pH, and temperature, to which inoculum is added), which gave rise at that time to several resistant/tolerant Guatemalan and Mexican rootstocks. We also screened using pots filled with naturally infested soil. The plant material tested was: (a) West Indian seeds from the original, local population; (b) seeds of the Guatemalan x West Indian hybrid Gema, taken from 20-year-old local trees growing in highly infected groves; (c) seeds of the salt-resistant Israeli West Indian cv. Maoz. Also tested for comparison purposes were seeds of Mexican and Guatemalan avocados. Inoculations were done using the P. cinnamomi A2 strain, taken from proved diseased trees found in the Canaries. At present, one West Indian plant has survived the selection in a first trial, while 38 have survived in a second trial. One Mexican plant (cv. Topa-Topa) has survived in a third trial. Resistant material is now being reproduced both by tissue culture and by the etiolation technique.     

Negative Disease Forecasting Prediction of the Disease-free Period of Irrigated Potatoes1

Although correlations can be established between the microclimate of an irrigated field and the macroclimate, their use is limited because the timing and duration of irrigation is not known beforehand, and it is practically impossible in consequence to predict crop microclimate in an irrigated crop. Nevertheless, a strong link can be found between some elements of the macroclimate and the disease-free period, for example for Phytophthora Infestons on irrigated potatoes. Temperatures of 10–12°C, or 18–20°C, early in the season are associated with long disease-free periods of 60–70 days. At intermediate temperatures, the disease-free period is shorter but cannot be predicted exactly. Such empirical relations can be used to provide « negative » forecasts, i.e. the expected duration of the disease-free period, and so avoiding unnecessary spraying.