Phytophage models
Informative and operational guide on forecasting models and DSS software related to the development of different pests.
Prediction model: Corn rootworm
Introduction
The corn rootworm, Diabrotica virgifera, is an important pest beetle of maize. Originally present in America, it is capable of attacking maize plantations, causing serious economic damage, especially in monoculture farms. The beetle overwinters in the form of pale yellow eggs that can be found in the soil at a depth that depends on the grain size of the soil. The looser the soil (sandy), the more the female insect lays her eggs on the surface. The maximum depth is 15 cm. In early spring, (generally starting in May), there is the metabolic recovery of the insect with the hatching of the earliest eggs. The larvae move from the oviposition sites and feed on the maize roots, causing significant structural damage that is evident later (generally between June and July). The damage is severe if the development of the insect is not treated and stopped in time and is caused by the formation of nutrient galleries within the roots with fatal consequences for the plant depending on the level of root growth. The ability of adults to cause damage even on resistant transgenic maize varieties, the ineffectiveness of crop rotation on the control of the pathogen, and the adaptation of some larval biotypes of the insect to specific protein toxins emitted by resistant varieties (hybrids) have triggered research and the development of predictive models for controlling the insect.
The WiForAgri model is able to predict the development of the insect in its different phenological stages. However, the prediction of the egg stage was omitted from the model, due to the difficulty of obtaining meaningful field data to validate the prediction of this stage.
By simulating the temporal evolutions of the different stages of development, and in particular the permanence of the larvae, the farmer will be able to
Rationalise the practice of early-early sowing with respect to the window of presence of larvae in the soil identified by the model.
Set larvicide treatments (geo-insecticides) in the period of greatest vulnerability.
Improving the effectiveness and decreasing adulticide interventions, which are particularly harmful to the environment (co-toxicity with pollinators), by predicting the cumulative emergence curve of adults.
The purpose of the Diabrotrica model is to provide the development status of the insect in an integrated manner, making it possible to predict and monitor the time windows of the different phenological phases of the insect. On the basis of this information, the grower will be able to better plan and manage control techniques and timing of phytosanitary defence.
Input data
Maximum and minimum daily (air) temperature
Cultivated maize hybrid (FAO Class)
Observed date of emergence of maize seedlings.
Latitude of location (given by system)
Output data
Percentage distribution curves of the pre-adult phenological stages: 3 larval stages, pupa stage (Davis submodel).
Percentage distribution curve of the adult phenological stage (Stevenson sub-model).
Bibliographic references
The model was developed by our development team with regard to the accessibility and availability of the input data and the adaptability of the model for the areas of interest. In particular, in order to have a complete forecasting system integrating all stages, two complementary models were chosen and implemented: the Davis (1996) model, able to predict the development of the pre-adult stages (larva and pupa) and the Stevenson (2008) model, able to describe the adult stage. The Stevenson model, based on the concept of degree days, was chosen for the low error demonstrated during validation in the original study (80% of predictions with an error <1.7 days). It was developed and validated over a period of ten years and is flexible enough to adapt both to different farm agronomic practices (different planting date, different seedling emergence date, different maize hybrids) and to different annual climatological trends. The Davis model, also based on degree days, was developed on data collected during a three-year period (1991-1993) within 11 sample locations. This model provides a prediction for the pre-adult stages and has shown good prediction accuracy for all four stages (3 larval stages + pupa stage) (mean square R = 0.86).
Coefficient of determination R square: Values equal to 1 indicate a perfect correlation between simulation and observation points. Values close to 0 a zero correlation.
Our modelling development team validated the calculation algorithms of the Davis and Stevenson models on some sample farms in the Veneto and Friuli-Venezia-Giulia regions. Within the validation campaign, the following activities were carried out:
The applicability of the model within the climatic subtype of North-East Italy and to the different maize cultivars widespread in the territory was verified.
The adaptability of the model to the local Diabrotica virgifera biotypes, which share the same genetic make-up, was verified.
Calibration of the model's calculation algorithms was performed on the different maize hybrids and different maturity rates (FAO Classes), sensitising it and improving its forecasting reliability.
The results of the good accuracy of the model from a forecasting point of view on some areas of Friuli Venezia Giulia are shown in Figure 3 below:
Description of the model
The WiForAgri model for Diabrotica virgifera uses linear equations to calculate degree days, which represent the daily heat input above the lower threshold temperature of phytophagus development. The calculation of degree days starts from the so-called biofix date, i.e. the date of metabolic activation of the phytophage. The degree days accumulated during the day are used by the model as an input variable for the determination of the daily development rate of the phytophage, i.e. the percentage in which a certain stage (larva, pupa, adult) has developed, completed its maturation and moved on to the next stage.
In the adult model, the biofix date, corresponding to the start date of accumulation of degree days, coincides with the observed emergence date of the maize seedlings. This information must be entered by the user as input data (see input data list).
For the correct calibration needed in the adult prediction model, the model will ask the user for the observed emergence date of the maize seedlings. This data is necessary at the start of the simulation for the prediction of the adult stage.
In the model of the pre-adult stages (larva, pupa), the determination of the biofix date is instead carried out by calendar, depending on the photoperiod and climatic characteristics related to the latitude of application of the model.
Through the calculation algorithms for the different development stages of Diabrotica, the model is able to determine, on a daily basis, the following output data
% emergence (first larval stage) and % development of the three larval stages.
% development of the pupa stage.
Start, evolution (%) and end of adult flickering.
The model uses linear equations to calculate degree days that take into account the two thermal thresholds of development, lower and upper, below and above which phytophage development stops, respectively. The thermal thresholds were determined for each stage (first larval stage, second larval stage, third larval stage) according to bibliographic values and then refined through a statistical regression procedure on the model development sampling data.
Graphical representation
Stage development is expressed as output data in two forms:
% cumulative stage emergence (0-100%), i.e. the percentage of individuals that passed within a given developmental stage (Figure 4b)
% relative stage emergence, i.e. the percentage of individuals passing within a given developmental stage during the day (Figure 4a)
Example: % adult cumulative emergence = 50% --> Means that 50% of the adults have fledged on the current day. % adult relative emergence = 10% --> Means that on the day 10% of the adults have flickered.
Figures 4a,b above show the graphical representation of the evolution of the different stages of the phytophage in relation to the testing of the model at a location in north-eastern Italy (Castions del Friuli, FVG) in the year 2016. Through the relative % emergence curves it is possible to observe the evolutionary development of the model, i.e. the start and end date of each phytophage stage, as well as the % development (rate of development).
The data provided by the DSS for Diabrotica virgifera, such as onset, evolution, peak and end, as well as the time window of presence of the larvae in the soil can be used by the decision-maker in order to rationalise agronomic control practices (early-postponed sowing) as well as to concentrate treatments in the periods of greatest phenological vulnerability of the phytophage, saving the product and protecting their arable production.
Prediction model: Corn borer
Introduction
The Maize borer, Latin name Ostrinia nubilalis, English name European corn borer (ECB), is an extremely polyphagous species (over 250 different vegetation species potentially host the insect) that is a harmful phytophagous insect of maize, which is mainly widespread in Canada, North Africa, Northern Europe, China and Russia.
The development of the insect is divided into four main phenological stages, egg, chrysalis, adult and five larval sub-stages. Overwintering takes place in the form of a mature larva (fifth larval stage) inside the stems (culms) left over from the previous year. In spring, adult flickering is observed and damage to leaf sheaths, ears and caryopses, as well as to culms, is caused by second-generation larvae.
Normally, in northern Italy the insect is rarely uni-voltine, and the number of generations may vary from one to five depending on the climate and genetic factors of the biotype present (Alma and Lessio, 2005). The deleterious action due to the undermining activity of the European corn borer is not only summarised in the economic loss of product, but in the increased risk linked to mycotoxins caused by fungi of the genus Fusarium and Aspergillus, which find a favourable environment for their development in the caryopses damaged by the phytophage. Mycotoxins are known to be toxic and dangerous carcinogens both to primary consumers (animals) and to humans, who eat milk and animal by-products fed with contaminated cereals. This public health problem has been acknowledged by the European Commission, which, through EU Regulations (EU Reg. 1881/2006; EU Reg. 1126/2007), has established toxicity thresholds, and by the MIPAAF, which has activated research projects and guidelines for toxicological control in the Regions and Autonomous Provinces.
Through the use of the forecasting tool for the Corn Borer of the WiForAgri platform the farmer will be able to observe the evolution of phenological stages and improve the logistics and effectiveness of treatments. Chemical treatment is in fact still the predominant strategy used for field control, and extensive damage can be prevented by striking the insect at a specific phenological stage of first generation.
Input data
Maximum and minimum daily (air) temperature
Latitude of location (system data)
Output data
Percentage distribution curves of all developmental stages of the first 3 generations (larval stages, pupa, adult, flickering, eggs)
Bibliographic references
The maize corn borer model integrated within the WiForAgri platform is structured on the compartmental system approach of G.C. Brown (1982), a generalised system applicable to a wide range of holometabolous insect species. By means of careful bibliographic research on the insect's rate of development based on daily thermal trends and a validation phase with traps in Veneto and Friuli Venezia Giulia, our development team first implemented and then calibrated Brown's compartmentalisation system, adapting it to the genomic characteristics of the Pyralid population present on the territory. In order to improve the predictive reliability of the model, an improved non-linear equation was introduced for the calculation of degree days, as opposed to the classical equation of Arnold (1960), which calculated them linearly as the difference between the mean temperature of the day and the lower threshold temperature of development.
With the aim of providing the best forecasting indications for corn borer infections affecting our territory, an activity was carried out to calibrate the bibliographic thresholds of degree days (Bessin, 2013) of the beginning and end of the insect's development stages, referring to the three annual generations (overwintering generation, first and second generation). The introduction of more realistic threshold values, combined with the improved equation for calculating degree days, yielded significantly more accurate results than the classical model formulation (Arnold equation) and made it possible to contain the prediction error within the reliability range (mean prediction error < 2 days) in the areas of application.
Description of the model
The WiForAgri model for the corn borer uses an improved non-linear equation for the calculation of degree days, which will be used as input data in the compartmental system to simulate the insect's developmental progress within the different developmental stages. This equation for calculating degree days introduces the concept of the three cardinal temperatures (lower, upper and optimum) specific to the species treated (Ostrinia nubilalis), i.e. it calculates degree days proportionally as the average air temperature approaches the optimum development temperature (non-linear approach). As the temperature approaches the insect's upper and lower threshold temperatures, and as the insect gradually moves away from the optimum temperature, the insect's accumulation of degree days slows down.
The degree days accumulated daily by the model are used to estimate the developmental status of the phytophage, i.e. the percentage (%) of maturation of the individual developmental stages (egg, five larval stages, chrysalis stage, adult stage). The compartmental system uses the pre-set degree-day thresholds at which the beginning and end of each developmental stage occurs to construct the developmental curves of the individual stages. In order to determine the end of the diapause and the resumption of metabolic activity of the Pyralid specifically for the area of application, corresponding to the start date for calculating degree days (biofix date), the model uses a mixed calculation approach that considers both the attainment of a certain number of daylight hours (photoperiod) and the thermal development of the day.
The model is able to determine, on a daily basis, the following output data: cumulative percentage of phytophage entry/exit at the different developmental stages: egg, five larval stages, pupa, adult, adult flight.
The corn borer model distinguishes between two stages relating to the phenological phase of the adult insect: the first stage (referred to as adult) relates to the adult's emergence from the chrysalis envelope; the second stage to the adult's subsequent actual unfurling (referred to as adult flight in the model).
Forecast model: olive fly
Introduction
The olive fly (Bactrocera oleae, GMELIN, 1790) is a phytophagus belonging to the subfamily Dacinae. It is a carpophagous species (it only feeds on fruit, in this case olives) whose larvae, through their undermining activity, cause great damage to the olive crops in which they are present. The larvae dig tunnels inside the olive pulp (mesocarp) and once they have reached maturity they pupate directly inside the olives or, vice versa, they drop to the ground, burying themselves.
In the event of infestation, the economic damage is considerable, both for direct causes (destruction of the pulp, olive collapse, rotting) and indirect causes with defects in the final product (e.g. increased acidity of the oil and development of bad smells).
Among the climatic factors predisposing to the attack, in addition to the temperatures that influence the development and mortality of the insect, there is the environmental humidity since a sub-humid microclimate (such as that of Northern Italy) in addition to ensuring the vital needs of the insect makes the olives more turgid and therefore more susceptible to the attack of the larvae. Another significant climatic factor is the low winter temperatures that, by increasing the mortality of the population, act as a real indirect control mechanism of the numbers during the growing season. In fact, the insect overwinters as a pupa in the soil; in areas with mild winters it may spend the winter season as an adult or as larvae in unharvested olives. A harsh winter can decrease the size of the overwintering population by affecting the survival of pupae in the soil. The insect reproduces several times a year: in coastal (warmer) areas 3-4 generations while in inland (colder) areas 1-2 generations. Temperatures in autumn and up to harvest completion, if mild, can sometimes allow for a further generation.
The dependency relationship between the scale of winter temperatures (particularly those in the period from November to February) and the criticality of infections in the summer period was recently examined and demonstrated in a recent study conducted on a large area in the Tuscany region (Marchi et al., 2015). It appears that low temperatures in the winter period act as a limiting factor for the phytophage more than the high temperatures of the summer period.
The incidence of winter temperatures can determine up to 50% of the intensity of the attack during the productive season following winter. In addition to winter temperatures, other factors competing with the severity of the infection are mainly of two types:
biotic type i.e. the presence of predators and parasites of the pre-adult stages of the fly;
of the plant phytological type i.e. the plant variety used and the leaf density of the plants.
In general, early or small-fruited varieties are the least affected. In fact, early varieties ripen before the peak of infestation in the autumn, while small-fruited drupes, besides being less attractive to adults, increase the egg mortality rate in the summer because their flesh tends to heat up a lot. Varieties with large, spherical fruits rich in pulp, such as table fruit, are more attractive to female flies and could therefore be used as spy plants.
Another recent observation, yet to be confirmed by further studies, reports that agronomic techniques that cause a delay in the ripening of olives may also generate an asynchrony between the peak flight of adult flies and the presence of ripe fruit attractive for oviposition. A Portuguese study, for example, observed that an abundant nitrogen supply can delay fruit ripening resulting in reduced susceptibility of the fruit to fly attacks.
Input data
Air temperature
Air humidity
Rainfalls
Captures of males/trap/week (average figure of the different traps). Example: Trap 1= 10 adults. Trap 2= 4 adults. → Captures males/trap = (10 + 4) = 14/2 = 7 adults/trap
Captures of females/trap/week (average data of the different traps). pole).
Output data
Risk index (relative to male catches)
Risk index (relative to female catches)
The type of input implemented by the user (female catch data and/or male catch data) results in the generation of two equally valid alternative risk indices.
The risk indices generated as an output of the model are useful for the timely and effective planning of adulticide treatments (preventive control), which is also possible in organic farming with special products.
Curative treatments, which target the larvae, are more complex to manage and may therefore be less effective. Moreover, for organic farms there are no systemic products allowed that can penetrate the olive tissue and kill the larvae.
With regard to the display of risk indices, the start date is set in the calendar (1 July by default) or activated by the user when the phenological phase of 'core hardening' is entered (BBCH=75 or later).
Bibliographic references
The WiForAgri forecasting model on the olive fly is based on the studies initiated by Pucci et al (1991) based in the first instance on the insect's female capture data. The decision to implement a DSS software primarily based on this series of studies, compared to other competing European models evaluated, was determined by several strengths listed below:
The model showed positive results in several test areas in central and southern Italy and in south-eastern Europe.
The model combines high simplicity with good forecasting reliability.
The model is adaptable/calibratable to different types of terrain and micro-climate.
By combining the number of weekly catches (male or female) with the air hygro-thermal trend (monitored by weather stations in the field), the model is able to generate a risk index that objectively assesses the phytosanitary criticality of the olive grove and determines the need for treatment
The model has also been updated and revised over the years by the Primo Principio team in order to optimise the model's adaptability (ergonomic solution) and its reliability, accuracy and predictive capacity. The WiForAgri model, unlike many other models based solely on degree days (phenological models), estimates the risk level by combining the micro-climatic and phenological approach with field monitoring. The use of chromotropic traps (for sampling females) and/or pheromone traps (for sampling males) is therefore required/recommended, the captures of which should be uploaded to the forecast model on a weekly basis.
Description of the model
The DSS (decision support software), for the purpose of suggesting or not suggesting a treatment, considers a standard damage threshold of 10% determined on the basis of a cost-benefit assessment.
If no trap catches are reported to the model in a given week (e.g. due to forgetfulness), the estimated risk level will be lower than the actual risk level. In any case, even in weeks with no catches on the traps it is possible to observe a risk index value above the treatment threshold due to the micro-climatic pattern favourable to the development of the post-adult stages (eggs, larvae, pupae).
The standard threshold on the risk index is -1 for the risk index based on male catches, and 0.1 for the risk index based on female catches. When these thresholds are exceeded, a sms/email alert will inform the user of the cost-effectiveness of carrying out treatment as soon as possible.
An application example with AIPO Verona
The result of the model was compared with field data (trapping) and traditional calendar-based treatment practices historically used by olive growers.
Figure 2 shows the treatments that the olive growers carried out 'on schedule' (orange triangles): the first was carried out in mid-June, the second well beyond the risk threshold (hence late) and the last when, based on trap catches, there was no longer any risk of attack. The yellow rhombus indicates the adulticide treatment in accordance with the modelling indications.
In contrast to the olive growers' calendar, the model identified an appropriate time window where the farmer should have carried out the treatment. (weeks 4-5).
In conclusion, it was observed that the model made it possible to rationalise control treatments by identifying the optimal time window based on the actual risk linked to the presence of the phytophage and its ability and speed of reproduction.
Corrective effects introduced by the model
Improve the timing of treatments by highlighting the uselessness of the first and last treatment.
Improving the effectiveness of the treatment by striking the insect at its moment of maximum vulnerability, reducing its reproductive potential and its numerous stays in the olive grove.
Expanding the choice of phytoiatric timing by the olive grower who is forced to cadence treatments over long intervals of time to prevent the emergence of forms of resistance to treatments by the phytophage.
For further model validation studies, please refer to the attached publications (statistical correlation data between index trend and drupe infection level).
Bibliografia di riferimento
Castoro V., Pucci C. (1996). Applicazione di un modello statistico di previsione della gravità dell’infestazione di Bactrocera oleae (Gmel.) (Diptera: Tephritidae) nell’ambiente olivicolo materano: esperienze condotte nel biennio 1994-1995 – Atti “Giornate Fitopatologiche”, Numana (AN) 22-24, 1: 505-512.
Crovetti A., Quaglia F., Loi G., Rossi E., Malfatti P., Chesi F., Conti B., Belcari A., Raspi A., Paparatti B. (1982). Influenza di temperatura e umidità sullo sviluppo degli stadi preimmaginali di Dacus oleae (Gmelin). Frustula entomologica, n.s. 5: 133-165.
Di Lena B., De Laurentiis G.,. Di Minco G., Di Giovanni R., Angelucci S., D’Ercole L. (1999). Verifica di un modello statistico di previsione dell’infestazione da Bactrocera Oleae Gmel. nei diversi ambienti olivicoli della regione abruzzo. Atti Giornate di studio su “Metodi numerici, statistici ed informatici nella difesa delle colture agrarie e forestali.” Sassari 19-22 maggio 1999.
Marchi S., Petacchi R., Guidotti S., Ricciolini M. (2015). Mosca delle Olive: Un modello previsionale per salvaguardare la qualità. pp. 66-70. L’Informatore Agrario n. 6/2015.
Matassa P., Antenucci F., Di Lena B. (1993). Verifica dell'applicabilità di un modello climatico per la previsione degli attacchi della mosca dell'olivo nel comprensorio vastese. Atti Convegno Nazionale "Protezione delle colture- osservazioni, previsioni, decisioni" Pescara 7-8 ottobre 1993.
Pucci C., Ballatori E., Forcina A. (1979). Soglia economica d’intervento per trattamenti diretti contro gli stadi preimmaginali del Dacus oleae (Gmel.) – Notiziario sulle Malattie delle Piante, 100 (26): 121-161.
Pucci C., Forcina A., Salmistraro D. (1982). Incidenza della temperatura sulla mortalità degli stadi preimmaginali, sull’impupamento all’interno delle drupe e sull’attività dei parassiti del Dacus oleae (Gmel.) – Frustula Entomologica, 4: 143-155.
Pucci C., Paparatti B. (1994). Prospettive di controllo guidato della Bactrocera Oleae (Gmel) mediante l’applicazione di un modello statistico di previsione della gravità dell’infestazione. Atti del Convegno “Lotta Biologica ed Integrata per la difesa delle colture agrarie e delle piante forestali”, Ferrara 24-25 ottobre 1994: 209-211.
Pucci C., Spanedda A. F., Paparatti B., Speranza S. (2006). Modelli di previsione della gravità dell’infestazione da Bactrocera oleae (Gmel.) (Diptera, Tephritidae), pp. 23-36. In: Medunarodna Manifestacija o Mastini i Maslinovom Ulju “Tekuće zeleno zlato Istre”, Croatia.
Raspi A., Conti B., Crovetti A. (1993). Verifica dell'applicabilità di un modello previsionale dell'andamento delle generazioni daciche in oliveti dei Monti Pisani. Atti Convegno Nazionale "Protezione delle coltureosservazioni, previsioni, decisioni" Pescara 7-8 ottobre 1993.
Spanedda A.F., Pucci C. (2004). Performance comparison between two forecasting models of inftestation caused by olive fruit fly (Bactrocera oleae Rossi). POMOLOGIA CROATICA, 12: 193-202.
Speranza, Stefano & Pucci, Claudio & Iannotta, Nino & Duro, Natasha & Jaupi, Alban & Thomaj, Fadil & Paparatti, Bruno. (2013). Application of a statistical forecast model on the olive fruit fly (Bactrocera oleae) infestation and oil analysis in Albania. Bulletin of Insectology. 66: 309-314.
Prediction model: Vine Moth
Introduction
Lobesia Botrana or the Grapevine Moth, is an insect that is widespread throughout Italy, whose seasonal presence varies depending on local microclimatic conditions, and in particular on the hygrothermal conditions that regulate its interspecific relations with its older sister, E.ambiguella (Grapevine Moth), in the different wine-growing areas.
The insect has an accentuated presence in the vineyards of southern Italy (and the islands), where the higher temperatures cause a shortening of the generational cycle: whereas in the vineyards of north-central Italy the Lobesia tends to carry out two generations, in the vineyards of southern Italy the generations usually range from three to four and the damage and consequently the phytosanitary measures are greater. With the upward trend in temperatures caused by global warming, the insect finds temperature trends more favourable to its development, and it cannot be ruled out that its presence will become more serious even in less affected wine-growing areas.
Overwintering at the chrysalis stage under the rhytidome, the adults of the first-generation insect flicker in spring (April-May) and lay their eggs on the flower clusters (anthophagous generation). The larvae that emerge cause modest damage to the flower buds, which becomes conspicuous during second-generation larval activity, with with withering and browning of the developing berries and loss of production. The third generation behaves like the previous one, attacking the berries. In these cases, the generation appears at the end of summer, when the berries are ripening, and is even more dangerous due to the fungal complications (Botrytis and acid rot) that can develop on the wounds caused by the larvae.
Phytosanitary interventions are based on the principles of integrated pest management and the setting of certain intervention thresholds: on the basis of the Emilia-Romagna Region's integrated pest management specifications, the thresholds are the presence of eggs or penetration holes in usually infested vineyards or 5% of infested bunches in vineyards that are not usually infested (second generation); finally 5% of infested bunches (third generation). The products used in organic farming are preparations based on Bacillus Thuringiensis (variety kurstaki), which acts by ingestion on lepidopteran larvae in their early life stages.
Bibliographic references
Provincial technical bulletins, which are useful in the identification of time windows for pesticide action, are usually based on forecast models, which have a number of advantages over the use of traps in the field. The Primo Principio staff developed the model for Lobesia botrana for farmers who, in possession of a quido-wiforagri agro-meteorological station, want to obtain an additional forecasting system on their vineyard, which can supplement and refine the indications given by the provincial technical bulletins for integrated pest management. The latter have the structural disadvantage of being based on radar-type weather data, which are less precise than capillary information.
Based on the most recent modelling of the original distributed delay model (Manetsch, 1976; Vansickle, 1977), in its final Time Varying Delay (TVD) version, the model was developed and integrated in-house by adding to the basic mathematical structure unresolved modelling aspects for the new application locations, such as the oviposition process, the influence of humidity on egg hatching, the speed of stage development and finally the effect of larval diet on the acceleration of larval and adult development. Some parameters were also made editable by our agro-mathematical managers (compared to the static model design) in order to make the new model rapidly adaptable to the new field realities represented by the vineyards being monitored.
The integration of the laboratory observations, carried out on the basis of temperature-controlled breeding, of bibliographic origin, with the field observations, based on the flight of the adults, allowed the model to improve its forecasting sensitivity, starting from the areas of North-Eastern Italy, the territory in which it was developed, with the Agri-CS project financed through SubMeasure 1.2 of the 2014-2020 Rural Development Programme of the Friuli Venezia Giulia Region.
Lastly, given the difficulty of determining the start time of the model, corresponding to the date of emergence of the adults from the overwintering chrysalises, with good precision for the locality under consideration, a sub-model was added to simulate the metabolic process of diapause of the Moth population, capable of giving the optimal start time offset of the model at the beginning of the season, based on the temperature and photoperiod trends to which the overwintering chrysalises in the vineyard are subjected. The presence of self-supplying agroclimatic stations makes it possible to sample data even during the winter season and to implement the diapause sub-model during that season.
For more references click on the following blog: https://www.wiforagri.com/2020/06/25/contrastare-la-tignola-della-vite-con-il-modello-previsionale-fenologico/
Input data
Average daily temperatures (collected from quido-wiforagri station)
Average daily relative humidity (collected from quido-wiforagri station)
Location coordinates (identified by the system during station installation)
Output data
Reference generation number (example: 2)
Phenological curves of all stages
Number of oviposited eggs on the day
Number of mature larvae on the day
Number of pupae matured on the day
Number of adults flickered on the day
The number of eggs, larvae, pupae and adults visualises the phenological development of the Lobesia population. However, the number represented is normalised and does not represent the actual number of insects found (e.g. in one hectare). Treatments are based on certain phenological stage onset (or peak) times
Description of the model
The Distributed Delay Model (DDMs) (Manetsch, 1976; Vansickle, 1977) simulates, through a series of mathematical equations with the same structure but different parameterisation, and cascaded between them, the passage of individuals of the Lobesia population within a series of 'k' compartments, which in turn represent the phenological developmental stages of the insect (egg, larva, chrysalis, adult).
By means of the mathematical functions of the diapause sub-model, based on temperature and photoperiod trends during the winter season, the model is able to estimate the adult flickering curve at the beginning of the insect's annual life cycle, and thus simulate the number of individuals beginning their development on the basis of environmental conditions. From the adult flicker curve, the model tells the user the number of adults of the same age that flicker on a given day and calculates their ageing rate, based on daily temperature trends. Bieri's (1983) algorithm, parameterised on the basis of temperature and larval diet, simulates the oviposition process of the female on the flower buttons (first generation) or on the berries (from the second generation onwards) of vine plants, thus indicating to the user the egg presence curve on the plants. At this point, the egg-laying individuals on the various days will represent the various cohort groups (or cohorts) that will develop and which will be placed within the compartmental system visualised in Figure 1. This system contains a series of 'h' boxes representing the various sub-stages in which the various cohorts (or individuals of the same age) must mature in order to emerge at the next phenological stage. Thus, macroblock k=1 represents the maturation process of the eggs, macroblock k=2 represents that of the larvae, macroblock k=3 that of the pupae, and so on. Back at the adult stage, the model recalculates the oviposition process of the female and so on for all the different generations, synchronising with the maturation of the individuals in the vineyard.
The residence time of individuals within the 'h' sub-stages of development was modelled through a series of 'development rate' equations that based on temperature and larval diet estimate the maturation rate of individuals. The daily mortality rate (or attrition) of individuals was also integrated within the model and is based on daily temperature values, where at the 'extremes' of the species' liveability temperature there is a higher mortality of individuals. Through the quantification of the development times and mortality of individuals of the same age (cohorts), the model is able to determine the number of larvae, pupae and adults that emerge in a given day from the 'k' macro-blocks, thus constructing the phenological curves on which treatments can be applied.
Graphical representation
Below is a representation of the graphical interface of the model:
This interface reports the (normalised) number of individuals present in the egg, larva, chrysalis and adult stages on the various days. The various phenological curves will give the user fundamental indications regarding the administration of pesticides. On the basis of these, it will be possible to act on the time window of larval presence by targeting the insect with Bacillus Thuringiensis-based preparations or with other types of products.
The use of SMS / Email / Telegram alerts will directly inform the user of the attainment of a certain threshold of population presence at a certain stage, thus automating the warning system
The number of individuals at a given phenological stage represents a fictitious, i.e. unrealistic number of the actual Lobesia population density in the vineyard. Through the use of pheromone traps, the user can independently assess the extent and severity of the insect infestation. The usefulness of the model remains in the prediction of the start, peak and end times of the different phenological stages.
Graph of phytosanitary indications
Reference bibliography
Alilla R., Severini M., Pesolillo S., 2005. Modello a ritardo distribuito a temperatura variabile per la simulazione dello sviluppo ontogenetico in stadi giovanili di popolazioni peciloterme. Rivista Italiana di Agrometeorologia 3: 30-33.
Alilla R., Severini M., Pesolillo S., Johan Baumgärtner., 2004. Fenologia della vite, e della Lobesia botrana (Lep. Tortricidae) nella zona dei castelli romani. Rivista Italiana di Agrometeorologia 3: 34-39.
Baumgärtner, J., Baronio, P., 1988. Modello fenologico di volo di Lobesia botrana Den & Schiff. (Lep. Tortricidae) relativo alla situazione ambientale della Emilia Romagna. Boll. Ist. Ent. “G. Grandi”, Università Bologna, 43: 157-170.
Baumgärtner, J., Gilioli G., Gutierrez, A, Ponti L., 2017. Climate warming effects on grape and grapevine moth (Lobesia botrana) in the Palearctic region. Agricultural and Forest Entomology,20 (2) :255-271. (https://onlinelibrary.wiley.com/doi/full/10.1111/afe.12256)
Baumgärtner, J., Gutierrez, A. P., Pesolillo, S., & Severini, M., 2012. A model for the overwintering process of European grapevine moth Lobesia botrana (Denis & Schiffermüller) (Lepidoptera, Tortricidae) populations. Journal of Entomological and Acarological Research, 44(1), e2.
D’Ercole L., Di Lena B., Giuliani D., Mazzocchetti A., Zinni A., 2012. Analisi di un decennio di voli di Lobesia Botrana nella Regione Abruzzo: Relazioni tra sommatorie termiche e dinamica dei voli. ATTI Giornate Fitopatologiche, 1: 413-420.
Logan J. A., Wollkind D. J., Hoyt S. C., Tanigoshi L. K., 1976. An analytical model for description of temperature dependent rate phenomena in arthropods. EnvEntomol5: 1133-1140.
Lozzia, G.C., Vita G., 1987. Preliminary notes on application of a predictive model for Eupoecilia ambiguella (Hbn.) and Lobesia botrana (Den. & Schiff.) flight in Lombardia (Italy) in relation to temperature. In Cavalloro R., 1989 (ed) - Influence of the environmental factors on the control of grape pests, diseases and weeds.
Riedl, H., 1983. Analysis of codling moth phenology in relation to latitude, climate and food availability, pp.: 233-252.
Roditakis E., Karandinos M., 2001. Effects of photoperiod and temperature on pupal diapause induction of grape berry moth Lobesia botrana. Physiol. Entomol. 26: 239-340.
Zinni A., Giuliani D., Di Lena B., D’Ercole L., Mazzocchetti A. 2012. Analisi di un decennio di voli di Lobesia Botrana nella Regione Abruzzo: Relazioni tra sommatorie termiche e dinamica dei voli
Fonti Web:
Alilla R., Severini M., Pesolillo S. 2005. Time varying distributed delay model for simulating the ontogenetic development in juvenile stages of Poikilothermic popolations. Fonte Web: http://www.agrometeorologia.it/documenti/rivista10_3/Alilla.pdf
ARPA Sardegna. Scheda informativa: Tignoletta della vite. Fonte Web: http://www.sar.sardegna.it/documentazione/agro/tignoletta.asp
Baumgartner J., Baronio P. 1988. Modello fenologico di volo di Lobesia Botrana Den.&Schiff. (Lep.Tortricidae) relativo alla situazione ambientale della Emilia-Romagna. Fonte Web: http://www.bulletinofinsectology.org/pdfarticles/vol43-1989-157-170baumgartner.pdf
Bevione D., Morando A., Morino G., 1990. Prove di controllo delle Tignole della vite con prodotti tradizionali e regolatori di crescita. Informatore Agrario 16/90: 141-145. Fonte Web: http://www.viten.net/files/b0e/b0ead0c9495be49c02ebf8a66a6619bc.pdf
Bodini A., Pasquali S. (Servizio Fitosanitario Regione Veneto). 2016. Modelli previsionali: la tignoletta della vite. Fonte Web: https://www.regione.veneto.it/c/document_library/get_file?uuid=fca329c3-e7bb-4763-b9ad-9751229fbc&groupId=10701
CRATI. Modello di sviluppo di popolazione di “Tignoletta della vite”. Fonte Web: http://www.crati.it/lobesia2.htm
CRATI. Modello per la Lobesia Botrana. Fonte Web: http://www.crati.it/relazione_modello.htm
Regione Emilia - Romagna. I MODELLI PREVISIONALI PER GLI INSETTI. Fonte Web: https://agricoltura.regione.emilia-romagna.it/fitosanitario/doc/prodotti-fitosanitari/Manuale-basso-impatto/documenti/parte-2-la-giustificazione-degli-interventi/6-i-sistemi-di-previsione-e-avvertimento/6-1-1-modelli-previsionali-per-gli-insetti/at_download/file/6.1.1%20Modelli%20previsionali%20per%20gli%20insetti.pdf
Regione Emilia - Romagna. Il modello MRV-Lobesia. Ultima modifica: 2018. Fonte Web: http://agricoltura.regione.emilia-romagna.it/fitosanitario/doc/previsione/insetti/il-modello-mrv-lobesia
Severini M., Alilla R., Pesolillo S., Baumgärtner J. 2005. Fenologia della vite, e della Lobesia Botrana (Lep.Tortricidae) nella zona dei castelli romani. Fonte Web: http://agrometeorologia.it/documenti/rivista10_3/severini.pdf
Prediction model: Vine Moth
Introduction
Eupoecilia ambiguella (or Clysia ambiguella) is the larger (in size and development) of the two grapevine moths. Compared to Lobesia botrana, which has a distribution area throughout the peninsula, loving the hot and dry climate, C.ambiguella is widespread in the cooler areas of northern Italy, i.e. in temperate and humid zones.
Also overwintering in the form of a chrysalis under the rhizidome and in the recesses of the vineyard (wooden poles, old vine stumps, etc.), it sees the over-wintering generation flicker in April-May and, as with Lobesia botrana, the female ovulates on the flower buttons of the vine plants. The insect's life cycle is then quite similar to Lobesia botrana with the difference that, as its development is slower (the insect reaches a larger size) and it likes cooler temperature zones, the Moth only carries out two generations per year (as opposed to three to four for the Moth). Very rarely, in warmer countries, there may be a third generation at the end of summer, but this is often partial and incomplete.
In vineyards where there is co-presence of the two species, the overlapping of the flight curves of the first two generations is generally observed, although it can sometimes be noted that the second generation of Clysia ambiguella is slower and later than that of Lobesia, and tends to coincide with the third generation of the latter.
Finally, the relative humidity conditions under which the eggs of the two species are laid on the flower buttons or berries of the parasitised plants directly influence the ability of the eggs to hatch, and this is also confirmed by field observations. In environments with relative air humidity between 40% and 70%, it has been observed that there is a higher percentage of Lobesia botrana eggs hatching. Conversely, in environments with higher air humidity, between 70 and 100%, there is a higher percentage of hatched eggs of Eupoecilia ambiguella. This would partly explain the different environmental optimum conditions required by the two species.
Lastly, the methods of control against C.ambiguella are the same as those already reported for the control of Lobesia botrana, especially in view of the fact that in most of the vineyards the phenological phases, and therefore the time windows for carrying out interventions, tend to coincide. Integrated pest management techniques, including the use of pheromone traps and Bacillus Thuringiensis-based preparations, again follow the intervention threshold values already seen for Lobesia botrana, what changes are only the type of pheromones, the traps used and the number of captures to determine the intervention threshold. Using grape sampling, the recommended intervention thresholds are 35-50% of infested bunches (out of 100 sampled) for the first generation (flower buttons), 5% of bunches infested with larvae or the presence of eggs or penetration holes for the second generation. Finally, for the last generation, if present, intervention with a threshold of 5% of infested bunches.
For more references click on the following blog: https://www.wiforagri.com/2020/06/25/contrastare-la-tignola-della-vite-con-il-modello-previsionale-fenologico/
Bibliographic references
The forecasting model for the grapevine moth, provided below, is developed by adapting the Time Varying Delay (TVD) model (Manetsch, 1976; Vansickle, 1977), in its final version, already previously developed for the grapevine moth, to the biological and phenological characteristics of the grapevine moth, i.e. to the different biometabolic response of the insect. Starting from bibliographic sources and on the basis of field observations carried out as part of the AgriCS project (for the development of phytosanitary models within the Friuli Venezia Giulia Region), our staff were able to construct a forecasting model for C.ambiguella, which could also be adapted to new territorial realities, always following the principle of maximum calibration elasticity of the model itself. The aspects that were calibrated and modelled are as follows: 1) The different hygrometric optimum of C.ambiguella compared to L. botrana was represented by introducing the Weibull function, capable of applying a coefficient to the egg birth rate on the basis of the species optimum in terms of humidity; 2) Lengthening of the adult ageing rate, on the basis of which the oviposition function of the female was lengthened 3) Introduction of a correction coefficient to the oviposition function of the female, according to which the number of eggs laid per female is lower in the grapevine moth than in the grapevine moth; 4) Use of a different set of development rate functions (compared to the grapevine moth) to better represent the development rate on the basis of daily temperature; 5) Finally, correction of the effect of the larval diet on that observed for the grapevine moth.
By modifying the aforementioned algorithms to the developmental response observed for the Grapevine Moth, we provide the user with a model adapted to represent the evolution of this species in the field. As with the Vine Moth, certain parameters have also been made editable by our agro-mathematical managers (with respect to the static conception of the model) in order to make the model rapidly adaptable to new field realities represented by vineyard types that are distant from each other, or where different biotypes of Vine Moth populations are present.
Input data
Average daily temperatures (collected from quido-wiforagri station)
Average daily relative humidity (collected from quido-wiforagri station)
Location coordinates (identified by the system during station installation)
Output data
Reference generation number (example: 2)
Phenological curves of all stages
Number of oviposited eggs on the day
Number of mature larvae on the day
Number of chrysalises matured on the day
Number of adults flickered on the day
The number of eggs, larvae, pupae and adults visualises the phenological population trend of C.ambiguella. However, the number represented is normalised and does not represent the actual number of insects found (e.g. in one hectare). Treatments are based on certain start (or peak) phenological stages
Description of the model
Distributed Delay Models (DDMs) (Manetsch, 1976; Vansickle, 1977) simulate the developmental dynamics of a cohort of a pecilothermic population in the juvenile (for animals) or phenophase (for plants) stages of the life cycle. These models belong to the category of Cohort - based Models (CbM), i.e. models that simulate the developmental dynamics of a group of individuals in a population at the same stage of the life cycle (cohort): more simply, they study groups of individuals that were born at the same time.
As can be seen from the structure shown above, a series of boxes (h=1, h=2, h=H) represent the different sub-stages of development through which the macroscopic developmental stage 'k' of the insect can be divided: egg, larva, chrysalis and adult.
Each box is represented by a mathematical equation that relates the flow of individuals exiting that box (y), with the flow of individuals entering that box (x) and with the parameters that determine the number and speed with which the individuals will pass that box. As already specified in the 'Model description' chapter of the Moth prediction model, these parameters that will determine the passage of individuals within a given box, or sub-stage of development, are air temperature, larval diet and stage k mortality (based on temperature).
By generating a multi-cohort population input to the system, in this case the number of eggs oviposited by the female moth, divided into 'groups' of individuals of the same age (oviposited on the same day), and placing the series of equations in cascade, where each relates the input flow to the output flow at the next substage, it will be simulated how a given population propagates through the different developmental stages.
The boundary conditions of the model will be a diapause sub-model, which estimates the flicker curve of the overwintering generation, and a female oviposition sub-model, which from the number of flickering adults determines the number of eggs oviposited during the day, and as an input cohort of individuals in Figure 1.
An alternative description of the structure of this model can be found in the chapter 'Model description' under 'Prediction model: Grapevine Moth' at the following link: https://wiki.wiforagri.com/wiki/v/wikieng/phytophage-models
Graph of phytosanitary indications
Reference bibliography
Agroatlas. Eupoecilia ambiguella (Hubner) - European Grape Berry Moth, Grape Berry Moth, Grape Bud Moth. Fonte web: http://www.agroatlas.ru/en/content/pests/Eupoecilia_ambiguella/
Gilligan, T. M. and M. E. Epstein. 2011. The European grape vine moth not found in California: Eupoecilia ambiguella (Hubner)., pp. 32-34. In Plant Pest Diagnostics Center Annual Report 2010. California Department of Agriculture, Sacaramento, CA.
Ibrahim, R. 2004. Biological control of grape berry moths Eupoecilia ambiguella Hb. and Lobesia botrana Schiff. (Lepidoptera: Tortricidae) by using egg parasitoids of the genus Trichogramma. Giessen: Köhler 2004.
Schmidt K., Hoppmann D., Holst H., Berkelmann-Löhnertz B. 2003. Identifying weather‐related covariates controlling grape berry moth dynamics*. EPPO Bulletin. 33: 517 - 524.
Schmidt K., Hoppmann D., Holst H., Berkelmann-Löhnertz B. 2003. Modelling the population dynamics of the grape moths L. botrana and E. ambiguella stages using age-structured models - the analysis of climate chamber experiments. n.d.
Scholten-Thoma, F. 1995. Optimierung eines Prognosemodells für Traubenwickler, Spezielle Untersuchung zur Flugaktivität und Eiablage bei Lobesia botrana (Schiff.) und Eupoecila ambiguella (Hübn.). PhD-Thesis, Mainz, 121 pp.
Prediction model: Scaphoid of the vine
Introduction
S.titanus makes one generation a year by overwintering as eggs under the rhytidome of vine plants, preferably on two-year-old or even older wood, singly or in groups of 3 - 6 (Vidano, 1964). This insect represents a strictly ampelophagous species: the biological cycle is carried out exclusively on species of the genus Vitis, mainly the European vine (Vitis vinifera L.). The eggs are fusiform in shape, creamy-white and approximately 1.5 mm long. There are 5 juvenile stages: the neanids (N1 and N2) have no winglets while the nymphs (N3, N4 and N5) have winglets; the adult male is about 4.5-5 mm long, while the female is about 5.5-6 mm long (sexual dimorphism).
Temperature is the environmental factor that most influences hatching dynamics and post-embryonic development. In the vineyard, the first N1 neanids appear between the first and second decade of May, while the adults are present from the beginning of July with a peak between the end of July and mid-August.
Direct damage is caused by all the stages sucking the plant sap with trophic stings: the reaction to this damage is very mild, with marginal necrosis and slight colour changes similar to the attack of other cicadas (e.g. Empoasca vitis). The most significant and serious damage from an economic point of view is the indirect damage due to the transmission of the flavescence dorée phytoplasma, which permanently compromises the infected vine. The transmission process comprises the following stages:
Acquisition: the vector feeds from the phloem of a plant infected with the flavescence dorée phytoplasma, which enters the insect's alimentary canal; this phase lasts from a few hours to a few days (4 - 5 days);
Latency: the phytoplasmas within the vector propagate and multiply for about 4 - 5 weeks;
Inoculation: the phytoplasmas have arrived in the insect's salivary glands and during suction bites are transmitted to new plants within the phloem stream.
In Italy, as well as in several European countries, the Flavescence Dorata phytoflasma and its vector insect Scafoideoides Titanus have been included among the phytopathologies subject to compulsory control, by Ministerial Decree No. 32442 of 31.05.2000. Through the Regional Phytosanitary Prescriptions, the areas at high risk of spreading (mainly abandoned vineyards) are identified, which will be subjected to compulsory containment treatments.
In order for the treatments to be successful and in accordance with the provisions of Directive 2009/128/EC on the sustainable use of plant protection products, the timing of the treatment must coincide with certain phenological phases of the insect. To this end, through the use of forecasting models, it will be possible to remotely control the phenological evolution of the insect, which essentially depends on temperature trends, and to plan the intervention in the field in advance.
Bibliographic references
The agronomic-mathematical staff of First Principle developed the Scafoideo titanus model using the framework already used for vine moth prediction models, i.e. the Distributed Delay Model (DDMs) (Manetsch, 1976; Vansickle, 1977). To this, the non-linear equation of Brière et al. (1999) was introduced to simulate the development rates of the pre-imaginal and adult stages. These equations are particularly significant as they simulate the non-linearity of the insect's developmental speed that is observed when approaching - receding from - ambient temperatures at the so-called lower and upper threshold temperatures. In particular:
The development of the insect accelerates-slows as it approaches-approaches the three stage-specific cardinal temperatures (lower, upper and optimum).
The different phenological stages of neanid (N1,N2), nymph (N3,N4,N5) and adult (A) are characterised by a different bio-metabolic response to temperature. From this point of view, a noticeably heterogeneous phenological development between the stages is to be expected.
The model was validated by collecting samples in wine-growing areas of the Piedmont Region (area delimited in red in Figure 1) and consisted of two project steps. The first step involved testing the development of the insect under controlled temperature conditions, the collection of related data and the parameterisation of the Brière equation. The second and subsequent step utilised the field data in witness vineyards for the purpose of calibration and field validation.
Input data
Daily average temperatures (collected from quido-wiforagri station)
Output data
Phenological curves of egg initiation and hatching evolution (Weibull equation)
Phenological curves of juvenile stage initiation and evolution
Numerosity of neanids (N1 and N2) and nymphs (N3, N4, N5)
Phenological curve of adult emergence and evolution
Description of the model
The Distributed Delay Model (DDMs) (Manetsch, 1976; Vansickle, 1977) simulates, through a series of mathematical equations with the same structure but different parameterisation, and cascaded between them, the passage of individuals of the Scaphoideus titanus population within a series of 'k' compartments, which in turn represent the phenological developmental stages of the insect (eggs, nestlings, nymphs, adults).
In the model, each box 'h' represents a sub-stage of development into which the main phenological stages of development of the insect can be divided, in this case for Scafoideo titanus the eggs, the juvenile stages N1,N2 (neanids) and nymphs (N3, N4, N5) and the adults.
Each box (sub-stage) is represented by a mathematical equation that relates the flow of individuals entering from the previous sub-stage to the number of individuals exiting the current sub-stage, which in turn will be the individuals entering the next sub-stage (cascading equation system). The mathematical relationship that correlates the inflow of individuals into the outflow of individuals within a given sub-stage is based on the temperature-dependent rate of development, which determines the time (or lag) with which individuals in the population stay within the sub-stage, and the also temperature-dependent mortality, which represents the number of individuals that will die within a given sub-stage.
By cascading the system of sub-stages, the model simulates the speed and mortality with which individuals propagate through the different evolutionary stages. Macroblocks k=1,2,3,K represent the various phenological stages of the insect, and by isolating the outflows from these macroblocks, it is possible to determine the number of individuals that emerge mature from a given phenological stage. In this case, the first macro-block k=1 will calculate the development of the first juvenile stage (neanid N1), the second macro-block k=2 that of the second juvenile stage (neanid N2), the third macro-block k=3 that of the third juvenile stage (nymph N3) and so on up to the adult stage.
In order to simulate the number of individuals entering the system (r0(t), Figure 2), i.e. entering macroblock k=1, of development of the first neanid phenological stage, the model uses the Weibull equation, which is able to estimate, on the basis of the accumulation of degree days, i.e. degrees above the egg development temperature, the percentage of eggs that hatch on a given day.
The number of eggs that hatch in a given day is the flow of individuals r0(t) that enter the compartmental system depicted in Figure 2. At this point, the rate of propagation of individuals in the different substages of the system is estimated through the development rate equations of Brière et al. (1999), which, on the basis of the daily temperature, estimates the time it takes individuals to move from one substage to the next.
Results of the calibration of the model
The model was developed, calibrated and validated within the Agri-CS project, financed through Sub-Measure 1.2 of the 2014-2020 Rural Development Program of the Friuli Venezia Giulia Region.
Within the project, the Primo Principio agronomic-mathematical development team went on to carry out a comparative procedure between the phenological curves simulated by the model and the field observations, provided by the technicians of the Regional Authority for Rural Development of FVG (ERSA). The calibration was carried out by: 1) setting a certain starting day for the degree day count of the Weibull function, which would minimize the differences between the number of hatched eggs observed and those simulated. 2) apply a corrective coefficient to the growth rate equations of Brière et al. (1999), which scaled the simulated development on the basis of the observed one.
The optimal calibration results of the Weibull function are shown below in graphic form
The validation, in graphic form, of the phenological curves of all the stages of the scaphoid, for a validation location (Moruzzo, Region FVG), is reported below.
Reference bibliography
Allen J. (1976). A modified sine wave method for calculating degree – days. Env. Entomol.
Alma A., Lessio F., Gonella E., Picciau L. (2014). Attuali conoscenze su Scaphoideus titanus.
Brière J.F., Pracros P., Le Ruox A.Y., Pierre J. S. (1999). A novel rate model of Temperature – dependent for Arthropods.
Falzoi S., Lessio F., Spanna F., Alma A. (2014). Influence of temperature on the embryonic and post – embryonic development of Scaphoideus titanus (Hemiptera: Cicadellidae), vector of grapevine Flavescence dorée. International Journal of Pest Management.
Falzoi S., Lessio F., Spanna F., Alma A. (2016). Real time forecast of the presence of Scaphoideus titanus Ball. In Piedmont.
Ferrari M. (2006). Fitopatologia, entomologia agraria e biologia applicata. Edagricole scolastico
Lessio F., Tedeschi R., Pajoro M., Alma A. (2009). Seasonal progression of sex ratio and phytoplasma infection in Scaphoideus titanus Ball. (Hemiptera: Cicadellidae). Bulletin of Entomological Research.
Jermini M., Morisoli R., Rigamonti I. E., Girgenti P., Mazzoni V. (2015). Fertility, longevity, oviposition dynamic and sex ratio of Scaphoideus titanus Ball.
Mazio P. (2014). PMScaTiLife: modello fenologico per Scaphoideus titanus come scaturito da quindici anni di osservazioni di campo in provincia di Reggio Emilia.
Rigamonti I. E., Jermini M., Fuog D., Baumgärtner J. (2011). Towards an improbe under standing of the dynamics of Vineyard – infesting Scaphoideus titanus leafhopper populations for better timing of management activities. Pest management science. 67. 1222-9. 10.1002/ps.2171.
Rigamonti I. E., Trivellone V., Jermini M., Fuog D., Baumgärtner J. (2014). Multiannual ainfestation patterns of grapevine plant inhabiting Scaphoideus titanus (Hemiptera: Cicadellidae) leafhoppers. Canadian Entomologist 146: 67-79.
Forecast model: Carpocapsa of the apple tree
Introduction
The apple codling moth (Cydia pomonella), whose discovery dates back to Roman times, is a phytophagous native to Central Europe whose presence originally corresponded to that of its initial host plant, the wild apple (Malus silvestris). Subsequently, the presence of the pest extended following the diffusion of the host fruit plants on new continents, in Asia, in North America and, sporadically, in the regions of the southern hemisphere. The codling moth of the apple tree is currently able to parasitize countless fruit plants in the pome fruit category, such as pear, apple and walnut, but also to a lesser extent some stone fruit, including plum and apricot, causing significant economic damage related to the productive value of the parasitized fruit plants. From a biological point of view, Cydia pomonella is a carpophagous lepidopteran which tends to complete from one to three generations a year depending on the territory of diffusion (in central-northern Europe even only one generation, in Mediterranean Europe generally three). It overwinters in the form of mature larva (5th stage) in the ravines of the barks of the host plants or at the level of the collar of the same in the ground. Once spring has arrived, there is the transition to the next phenological stage, the chrysalis and then subsequently, between April and May, the emergence of the adults occurs, which generally reaches its maximum point in the second half of May. Once mating is complete, the eggs are laid on the leaves. From these are born the larvae which, penetrating through the epicarp of the fruit, form a series of spiral galleries extending up to the center of the fruit which lead to the abscission and the fall of the latter. Additional economic damages are linked to the loss of the aesthetic quality of the product. Once the first generation is completed, if the environmental conditions allow it, the second generation larvae form the second generation chrysalises on the branches which in turn will give rise to a second flight between June and July culminating in the first half of July. The deposition of the eggs by the female now takes place directly on the fruits. The second generation larvae, depending on the climatic conditions, can enter diapause and end the cycle, or give rise to a third generation, whose emergence generally occurs from August to September.
Bibliographic references
In order to be able to model and predict the phenological stages of the Codling moth, the agronomic-mathematical staff of WiForAgri has developed a variable delay MRV model (already implemented for the Moths of the vine) which, through specific parameterization on the insect of interest, provides a robust decision-making tool capable of giving optimal treatment and monitoring indications. The decision to adopt an MRV model also for the apple codling moth was linked to the solidity and proven adaptability of this family of models in simulating the development and maturation of various phytophagous insects. All the bio-metabolic response functions and the setting parameters used in the WiForAgri model of the codling moth have been incorporated and adapted from the reference literature. In particular, the model is based on the breeding study of the phytophagous populations carried out at the Regional Observatory for plant diseases of Bologna (Emilia-Romagna Region) at the beginning and at the end of the 90s (Butturini et al, 1991) (Butturini and Tiso, 1999).
The rate/velocity functions of development of the different phenological stages, those of oviposition and mortality, and finally the number of unobservable stages of the model were parameterized and validated starting from observations of the demographic response of groups of individuals reared and stimulated at temperatures variables (constant photoperiod and humidity). A further calibration was carried out within a modeling development study of the "AgriCS" project, financed through Sub-Measure 1.2 of the 2014-2020 Rural Development Program of the Friuli Venezia Giulia Region, in which the curves were compared simulated by the model with those observed through field investigation, and a calibration procedure of the model functions was carried out, in order to provide maximum representativeness with respect to the various pedoclimatic scenarios of application.
Input data
Daily average temperatures
Output data
Reference generation number
Phenological curves of all stages
Number of eggs laid in the day
Number of larvae matured during the day
Number of chrysalises matured during the day
Number of adults emerged during the day
Description od the model
The Distributed Delay Model (DDMs) (Manetsch, 1976; Vansickle, 1977) simulates, through a series of mathematical equations with the same structure but different parameterization, and placed in cascade between them, the passage of the individuals of the Codling moth population to the inside a series of compartments "k", which in turn represent the phenological stages of development of the insect (eggs, larva, chrysalis, adult).
Through the mathematical functions of the diapause submodel, based on the trend of temperature and photoperiod during the winter season, the model is able to estimate the flickering curve of the adults at the beginning of the insect's annual life cycle, and therefore to simulate the number of individuals who begin their development on the basis of environmental conditions. From the adult flickering curve, the model indicates to the user the number of adults of the same age flickering on a given day and calculates their aging rate, based on the trend in daily temperatures. The algorithm of Bieri (1983), parameterized on the basis of the temperature and the larval diet, simulates the process of oviposition of the female on the flower buds (first generation) or on the berries (from the second generation onwards) of the vine plants, thus indicating all the user the curve of the presence of the eggs on the plants. At this point the individuals oviposed in the various days will represent the various peer groups (or cohorts) which will develop and which will be placed within the compartmental system visualized in Figure 1. This system contains a series of boxes "h" which represent the various substages in which the various cohorts (or individuals with the same age) must mature in order to emerge at the next phenological stage. The macroblock k=1 thus represents the maturation process of the eggs, the macroblock k=2 represents that of the larvae, the macroblock k=3 that of the chrysalises and so on. Returning to the adult stage, the model recalculates the female's oviposition process and so on for all the different generations, synchronizing with the maturation of the individuals in the vineyard.
The permanence time of the individuals within the "h" substages of development was modeled through a series of "development rate" equations which, on the basis of the temperature and the larval diet, estimate the speed of maturation of the individuals. The daily mortality rate (or attrition) of the individuals was also integrated within the model and is based on the daily temperature values, whereas at the "extreme" temperatures of livability of the species there is a greater mortality of the individuals. By quantifying the development times and mortality of individuals of the same age (cohorts), the model is able to determine the number of larvae, chrysalises and adults that emerge in a given day from the macroblocks "k", thus constructing the curves phenological conditions on which the treatments can be applied.
Reference bibliography
ARPA Veneto. VALUTAZIONE DELL’APPLICABILITÀ IN VENETO DEL MODELLO PREVISIONALE MRV-CARPOCAPSA PER CYDIA POMONELLA. Fonte Web: http://www.arpa.veneto.it/temi-ambientali/agrometeo/file-e-allegati/documenti/lotta-antiparassitaria/poster_Carpocapsa_Sassari.pdf
Butturini A., Checchetto F., Delillo I., Marchesini E., Tiso R., Zecchin G. (2009). Valutazione dell’applicabilità in Veneto del modello previsionale MRV-Carpocapsa per Cydia Pomonella. Atti del 12^ Convegno nazionale AIAM, “CLIMA E AGRICOLTURA: Strategie di adattamento e mitigazione”. in Rivista Italiana di Agrometeorologia, anno 14, n.2: 122-123.
Butturini A., Tiso R. (1999). Un modello fenologico per Cydia pomonella (L.) (Lepidoptera: Tortricidae) nella difesa delle pomacee in Emilia – Romagna. Frustula Entomologica 22: 113-120
Butturini A., Tiso R., De Berardinis E. (1992). Influenza della temperatura sullo sviluppo di Cydia pomonella (L.) (Lepidoptera: Tortricidae). Bolletino istituto di Entomologia G. Grandi Università di Bologna , 47: 123 – 134.
Damos, P. & Kouloussis, N. (2018). A degree-day phenological model for Cydia pomonella and its validation in a Mediterranean climate. Bulletin of Insectology. 71.
Tiso R., Boselli M., Butturini A., Bellettini L. (2001). Andamento dell’ovideposizione in campo di Cydia pomonella L. Informatore Fitopatologico , 51 (6): 33-39.
Last updated
