Decision Support includes projects whose aim is the development and promotion of decision systems useful in crop production and pest management. General areas of such work might include but are not limited to development or improvement of sampling or detection methods, quantification of crop loss, development or improvement of damage thresholds or action levels, risk assessment, economic or other evaluation of IPM programs, development or evaluation of expert systems or other computer models.
Development of monitoring programs for pests and biological control agents involving sampling decision rules, control action thresholds, or improved methods for quantifying pest abundance or potential crop loss can significantly reduce the riskiness of pest control, reduce the number of pesticide applications, and improve adoption of new practices. Control action threshold research should be based on the relationship of pest occurrence or abundance to measurable crop loss in order to ensure cost-effective decisions.
High priority is given to proposals that clearly define objectives; will produce programs that can be easily and directly used in the implementation of integrated pest management; will focus on a biological system that has had sufficient preliminary investigation to allow completion and validation of the program in a reasonable amount of time; and will involve principal investigators who have demonstrated experience in systems modeling or systems management.
New Projects Funded for 1997-98
SAMPLING FOR WEEDS
Precision agriculture: Comparison of weed seed and previous weed populations for prediction of subsequent weed populations. (Year 1 of 2; $18,760)
Principal Investigator: W. T. Lanini, Vegetable Crops, Davis
Objectives: Evaluate two sampling methods, weed seedbank (Elutriation-Germination method) and weed population sampling the previous growing season, alone or in combination, for predicting summer weed populations and their distribution within a field.
Analyze the weed populations at crop harvest and weed seedbanks during the winter, in terms of the species composition and spatial distribution, and determine the most reasonable sample size to describe such populations at a farmer's level of resolution.
BIOLOGY OF STING NEMATODE
Biology and life cycle of the sting nematode. (Year 1 of 2; $19,576)
Principal Investigator: J. O. Becker, Nematology, Riverside
Objectives: Describe life cycle of the sting nematode as observed in culture of excised corn roots.
Determine influence of various physical and chemical environmental factors on behavior and population development of the sting nematode.
PREDICTING PEAR SCAB OUTBREAKS
Modeling of pear scab, Venturia pirina, epidemiology and disease risk using an onsite weather monitoring system. (Year 1 of 3; $16,700)
Principal Investigator: D. Gubler, Plant Pathology, Davis
Objectives: Monitor pear scab pseudothecial development, ascospore maturation, ascospore release, and infection of plant parts from primary inoculum at two pear orchard field sites in Mendocino County, each with an onsite weather station with sensors for rain, leaf wetness, temperature, and relative humidity.
Refine the Spotts-Cevantes model for predicting pear scab ascospore release as well as conditions favorable for disease development, and incorporate these into an electronic format.
Test a treatment regime at one field site based on applying fungicides only when needed; i.e., when inoculum is viable and present in sufficient numbers and conditions are favorable for infection.
DEGREE-DAY MODELS FOR WEEDS
Using phenology prediction to improve weed management. (Year 1 of 2; $35,131)
Principal Investigators: J. S. Holt, Botany and Plant Sciences, Riverside; T. S. Prather, UC IPM Project, Kearney Agricultural Center, Parlier
Objectives: Quantify phenology of common annual weeds in two different locations to identify functional weed groups defined by similar threshold temperatures and degree-day requirements.
Determine the effect of crop presence on weed phenology and phenology predictions.
Test phenology predictions for weed management in field experiments and growers' fields.
STINK BUG PHEROMONES
Pheromone-based monitoring systems for major stink bug pests of agriculture in California. (Year 1 of 3; $30,000)
Principal Investigator: J. G. Millar, Entomology, Riverside
Objectives: Identify attractant pheromones for major stink bug pests in California, particularly species that infest a number of crops.
Continuing Projects Funded for 1997-98
GREENHOUSE SAMPLING FOR THRIPS
Development of sampling procedures for western flower thrips, Frankliniella occidentalis, in greenhouse flower production. (Year 2 of 2; $17,886)
Principal Investigators: M. P. Parrella, Entomology, Davis; B. C. Murphy, Entomology, Davis
Objectives: Determine the within- and between-plant distribution of western flower thrips on roses and chrysanthemums for common cultivars during the growing season and determine the optimal sample unit for estimating western flower thrips (WFT) abundance.
Develop and evaluate a practical, grower-oriented sampling procedure using sticky cards, plant samples, or a combination of the two for monitoring WFT to determine the need for control measures.
Estimate the relationship between WFT density on plants and the degree of feeding injury on foliage and flowers.
Verify effectiveness of the sampling procedures under commercial greenhouse conditions.
Summary of Progress: Effective IPM programs for the western flower thrips (Frankliniella occidentalis, WFT) in greenhouses have yet to be developed and lack of knowledge about monitoring procedures and action thresholds have been identified as an impediment to initiating IPM. The goal of this project is to develop commercial monitoring and action guidelines for WFT based on plant samples, sticky cards, or a combination of the two. During 1996, two commercial greenhouses were sampled for rose and two for chrysanthemums to obtain spatial distribution data on WFT for developing monitoring programs. Pesticide sprays at one chrysanthemum site prevented analysis of the results.
Analysis of the within-plant distribution determined the majority of thrips occur in terminal buds of nonblooming chrysanthemums relative to lower portions of the plants. For blooming chrysanthemums thrips were found to be more evenly distributed within plants. In nonblooming chrysanthemums, the spatial distribution of thrips were found to be evenly distributed within beds. Preliminary sample size estimates found examination of five or less terminal buds is sufficient to estimate economic densities of thrips. Results thus far suggest a cost-effective thrips monitoring program could be developed for chrysanthemums.
The within-plant distribution for roses also found the majority of thrips occur in terminal rose buds and flowers relative to lower portions of the plants. The spatial distribution of thrips were moderately aggregated within fresh cut rose buds. Preliminary sample size estimates determined examination of 20 rose buds is sufficient to estimate economic densities of thrips among homogeneous rose beds. In addition, our results found a significant relationship between thrips capture on blue and yellow sticky cards and thrips numbers in rose buds. These results suggest a cost-effective monitoring procedure could be developed for roses using plant samples, traps, or a combination of the two.
BROWN ROT PREDICTOR
Using detection of latent infections of Monilinia fructicola in stone fruits in California to predict brown rot at harvest and postharvest and reduce fungicides. (Year 2 of 2; $16,300)
Principal Investigator: T. J. Michailides, Plant Pathology, Davis
Objectives: Determine if incidence of latent infections of peaches and nectarines by Monilinia fructicola can be a predictor of brown rot of fruit at harvest and postharvest.
Use the predictor to minimize the use of preharvest fungicide sprays of stone fruit for the control of brown rot at harvest and postharvest.
Use the predictor to minimize the use of postharvest fungicide treatment of stone fruit for the control of postharvest brown rot.
Summary of Progress: If weather conditions are wet and cool in the spring this disease can attain epidemic levels, killing blossoms and rotting green and mature fruit as well as fruit in storage. In previous years' research, we obtained preliminary information that supports the hypothesis that we could be able to develop a simple method to predict fruit brown rot.
Because of the expected very low levels of brown rot in 1996, fruit from 7 nectarine and 13 plum orchards were sampled every 3 weeks and processed using the freezing-incubation method that we developed in our laboratory to detect latent infections caused by Monilinia spp. Due to the very unfavorable environmental conditions, levels of latent infections were in most cases zero and in no case did latent infections exceed 5%. As expected, the levels of disease recorded in the field just before commercial harvest did not exceed 4%, although postharvest brown rot ranged from 0 to 56%. Because of the extremely low and sporadic incidence of latent infections in 1996, no significant relationships could be determined between the incidences of latent infections of green fruit and brown rot of mature fruit at harvest and/or at postharvest.
In contrast to the very low incidence of latent infections in nectarines and plums, the incidence of latent infections in prunes was higher and there were significant relationships between latent infections and disease in the field and postharvest in a similar project with French prunes, supported by the California Prune Board. Additionally, based on the levels of latent infections, we identified three orchards with different levels of disease risk and applied fungicide sprays. The 2-day preharvest fungicide treatment was not applied as proposed because of the voluntary increase by the registrant of the fungicide iprodione (Rovral) of the preharvest interval (PHI) from 0 to 7 days.
There was a significant linear relationship between the incidence of latent infections of immature prunes collected from commercial orchards during late June and July with the incidence of brown rot in the field and after postharvest incubation of healthy fruit. Therefore, monitoring of brown rot latent infections in immature prune fruit can be a viable method to predict brown rot of prune at harvest or postharvest. Results from the preharvest fungicide trials showed that indeed neither of the two sprays were needed in an orchard predicted to have low risk of brown rot; one spray was sufficient to significantly reduce brown rot in the orchard predicted to have a medium risk of disease; and one late or two sprays were required for the greatest disease reduction in the orchard predicted to have high risk of brown rot. We plan to develop a similar method for nectarines, peaches, and plums.
Objectives: Evaluate the effect of a range of lygus bug populations on yield and quality of CB46 blackeye beans.
Evaluate current decision thresholds relative to timing of infestation.
Summary of Progress: Lygus bug is a key pest in cowpeas or blackeye beans. This insect can cause damage in two ways: by feeding on flower buds it can cause a loss of potential bean pods, or by feeding after the pods have formed it can reduce the quality of the beans. This project used 6' x 6'x 6' cages to cover two rows of beans (CB46) before the plants flowered. These cages provided protection from lygus bugs during the experiment while allowing a known number of insects to be introduced on plants. Just before flowering, lygus bugs were collected from the field sorted by sex. Into half the cages, 0, 20, 60, or 120 adult lygus males were placed for 27 days and then removed. Just after flowering, the other half of the cages received 0, 20, 60, or 120 adult male insects. Again, the insects were allowed to feed for 20 days and then removed.
We learned that the number of bugs introduced into the cages was less important than the time they were introduced. No difference in yield or quality was noted when the bugs were introduced before flowering but significant losses were caused if introduced after flowering. We suggest that this is related to the ability of the plant to compensate for early damage as well as an artifact of our single sex, limited time frame for feeding.
Projects that Ended in 1996-97
SILVERLEAF WHITEFLY IN SAN JOAQUIN VALLEY
Phenology, seasonal host selection, and biology of the silverleaf whitefly (Bemisia argentifolii Bellows & Oerring) in the San Joaquin Valley.
Principal Investigators: C. G. Summers, Entomology, Davis/Kearney Agricultural Center, Parlier; R. L. Coviello, UC Cooperative Extension, Fresno County; M. J. Jiménez, UC Cooperative Extension, Tulare County
Objectives: Determine the seasonal phenology and population dynamics, the seasonal host plant acceptability, and the crop host selection sequence of the silverleaf whitefly on selected hosts in the San Joaquin Valley.
Determine the biology of silverleaf whitefly and the role of selected crop and noncrop hosts in overwintering survival.
Establish thresholds values for nymphal populations causing broccoli white stalk and irregular ripening of tomatoes in the San Joaquin Valley.
Summary of Progress: Silverleaf whitefly crawlers move upward on plants. Experiments were conducted to determine if this movement was random or a response to light or gravity. Crawlers were positively phototactic. They moved up or down the petiole of cheeseweed toward a light source. Response was always toward the light and there was no response to gravity. Crawlers placed in a dark arena moved toward a light source. Crawlers illuminated with uniform light or kept in darkness moved about the arena at random. Crawlers maintained in darkness moved a shorter distances. Some minimal light intensity may be necessary to stimulate crawler activity.
Positive phototaxis contributes to silverleaf whitefly survival by enabling individuals to find suitable leaves for development higher on the plant. The preference of crawlers for the adaxial or abaxial leaf surface was investigated on cheeseweed leaves. In free choice experiments, crawlers exhibited a negative phototaxis, selecting the surface away from the light. When transferred directly to either leaf surface, crawlers remained and settled on the surface on which they were placed regardless of leaf orientation or illumination.
Following cotton defoliation, massive numbers of silverleaf whitefly adults move into surrounding citrus and deposit eggs on new growth and on weeds, particularly spotted spurge, in the understory. Citrus, at least in eastern Tulare County, appears to be the principal overwintering site of silverleaf whitefly. There is a direct relationship between silverleaf whitefly densities in cotton, broccoli, and other vegetables and their proximity to citrus groves. There is in inverse relationship between density in these crops and their distance away from the citrus belt.
There was a direct relationship between whitefly density and the percentage of tomato irregular ripening in both fresh market and processing tomatoes. Results for 1996 were similar to 1995 indicating that <25/leaf can cause substantial irregular ripening. Maturity is delayed by increasing whitefly density; a significantly higher percentage of green fruit (and lower percentage of ripe fruit) was harvested from high whitefly density plots.
RICE/WEED SIMULATION MODEL
Management-oriented simulation of rice growth and weed competition.
Principal Investigators: T. C. Foin, Environmental Studies, Davis; J. E. Hill, Agronomy and Range Science, Davis
Objectives: Continue experimental research on rice-weed interactions that (1) advances the search for the characteristic properties that are relatively independent of the identity of the weed species; and (2) provides the parameter values necessary for the simulation models.
Complete the revisions to the California Rice (CARICE) simulation model needed to adapt it for predicting weed impacts and to evaluate alternative strategies for weed control.
Summary of Progress: Final validation experiments were completed in the field and greenhouse. An herbicide experiment at the Tres Picos Ranch, also in Butte County, was done in which four harvests were made over the season. This will provide more data points over the growing season for validation than are available from previous experiments. This field experiment included the weeds Ammannia spp. (redstem) and Scirpus mucronatus (ricefield bulrush).
A greenhouse competition experiment between rice and Cyperus difformis (smallflower umbrellaplant) was repeated from 1995, with an additional mid-season harvest added. We also obtained access to a light bar this year and were able to measure canopy light levels in field and greenhouse experiments. These data will enable a meaningful comparison of model output to experimental results.
We completed the development of Water-Seeded Rice (WSRICE), a growth and weed management model for water-seeded rice. WSRICE combines empirical research on mechanisms of rice-weed competition, with state-of-the-art plant growth modeling techniques to simulate rice phenology, growth, and competition for light with weeds, WSRICE permits experimental manipulation of flooding date, mean water depth, seeding date, rice cultivar, rice seed rate, and herbicide application dates and types. Weed seed or plant density is also input into the model. WSRICE also allows experimental manipulation of the key management practices, perhaps with the exception of fertility, that growers use for weed control and rice productivity, linked in a practical manner to phenology. Validation, testing, and analysis of the model is underway. WSRICE will allow experimental evaluation of weed management strategies and optimization of herbicide application.
CITRUS THRIPS DECISION-MAKING
Risk assessment for fruit scarring by citrus thrips, Scirtothrips citri (Moulton), from soil and air temperatures before petal fall.
Principal Investigators: H. Schweizer, Entomology, Riverside; J. G. Morse, Entomology, Riverside
Objective: Develop a temperature-driven, phenological model for estimating the amount of fruit scarred by citrus thrips, which can be used to decide whether or not control measures are needed.
Summary of Progress: In spring 1995 and 1996, ground and air temperatures, citrus thrips populations, and tree phenology were monitored at a total of 11 navel orange sites in the San Joaquin Valley (from Woodlake to 30 miles south of Porterville). Originally we intended to use these field data for developing a rather rigorous mechanistic fruit-scarring model. However, because of the relatively low thrips scarring at all sites in both seasons (less than 6% of outside fruit scarred at all sites) this approach appeared less promising and we resorted to a black box approach.
Using historical fruit scarring and air temperature data, we developed regression models to estimate the amount of fruit scarring in a given year from temperature patterns before the end of petal fall. The models were tested against independent validation data. Assuming a very effective and inexpensive insecticide treatment (Baythroid, $45/acre; Baythroid reduces the proportion of severely scarred fruit always to 4%), an economic analysis was performed.
Our results suggest that using the predictions from our model instead of indifferently applying insecticides against citrus thrips in every season would result in increased monetary return per acre. In 3 out of 13 seasons, no treatments had to be applied. Our best model suggests that cool weather in early March and warm weather during bloom is usually associated with high citrus thrips scarring. Conversely, in seasons with high temperatures in early March and low temperatures during bloom, fruit scarring tends to be low.
Objectives: Determine and compare the effectiveness of combinations of reflectorized plant coatings, soil mulches, and the aphid alarm pheromone in delaying and/or reducing the development of aphid-borne virus diseases and subsequent crop loss in susceptible vegetable varieties.
Determine the extent to which these various methods and combinations repel aphids from plants thus delaying and/or reducing colonization and virus transmission.
Determine the effects of these treatments on earliness, health, yield, and economics of susceptible vegetable crops.
Summary of Progress: A complex of aphid-vectored virus diseases is causing increasingly heavy losses in many vegetable crops grown in the inland valleys of California. Five field experiments (two at UC field facilities; three on-farm with grower/collaborators) were conducted in 1996 in cantaloupe melon, cucumber, pumpkin, fresh market tomato, and zucchini squash to test the effectiveness of reflectorized, spray, and polyethylene soil mulches for management of aphids and aphid-borne virus diseases in San Joaquin Valley vegetable crops. Effects of treatments on seasonal dynamics and damage caused by the silverleaf whitefly (Bemisia argentifolii) also were determined. As originally proposed, implementation of the project in commercial fields was stressed during this final year of the project. Excellent yield responses from the experimental crops were obtained to the reflectorized mulches under moderate virus conditions. Aphid numbers and virus disease incidence on mulched plants were consistently lower than on nonmulched plants in all experiments, which was critical for facilitating normal flowering, fruiting, and yield. An added benefit not anticipated at the beginning of the project was the repellency of the reflective mulches to silverleaf whitefly, giving good control of silverleaf symptoms in curcubitaceous crops. No significant benefit from plant coatings or the aphid alarm pheromone was obtained on any of the crops.
A monitoring system for predicting tospovirus epidemics by sampling for infective thrips.
Principal Investigator: D. E. Ullman, Entomology, Davis
Objectives: Adapt the existing enzyme-linked immuno- sorbent assay (NSs ELISA) technology for use in greenhouse grown crops.
Determine an inexpensive and easy method for trapping thrips so they are adequately preserved for use in ELISA. Determine a trapping media that does not interfere with the ELISA test, but preserves the thrips.
Determine how long the NSs protein can be detected after death of the insect on the trap.
Refine the existing ELISA method for use on large numbers of insects.
Establish that the ELISA can detect naturally infected thrips under a field setting.
Summary of Progress: Tomato spotted wilt virus is transmitted by thrips and causes disease in up to 600 species of plants. Growers of floricultural and ornamental crops rank thrips and spread of tomato spotted wilt virus (TSWV) as one of their most serious unsolved problems. The goal of this project is to combine use of a new technique for detecting infective thrips with traditional sampling methods (i.e., yellow sticky cards) to identify and quantify TSWV vectors.
Infective thrips are quantified with a technique called enzyme-linked immunsorbent assay (ELISA) that allows detection of a protein that the virus produces when it multiples in thrips. Previous research showed that this technique allows separation of infective from noninfective thrips. The ultimate application of this technology, which was originally developed as a basic research tool, will be as part of a monitoring system that rapidly and reliably localizes sources of infective thrips and TSWV, allowing growers to predict epidemics and direct application of control measures.
Since initial funding in July 1996, we have determined that thrips trapped on yellow or blue sticky cards can be removed from these cards and successfully tested with ELISA. Infective thrips could be detected from cards held under greenhouse and field conditions for up to 2 weeks and shipped under standard US Postal Service conditions. Although not sufficiently advanced to draw conclusions, field experiments using this technology in conjunction with other predictive tools (i.e., TSWV indicator plants) are underway to test the applicability of the technology as part of the monitoring needed by growers.
Final Reports for Projects that Ended in 1996
GRAY MOLD PROTECTION FOR KIWI
Using levels of kiwifruit sepals and stem ends colonized by Botrytis cinerea to predict gray mold and reduce fungicides.
Principal Investigators: T. J. Michailides, Plant Pathology, Davis/Kearney Agricultural Center, Parlier; D. P. Morgan, Kearney Agricultural Center, Parlier
Summary of Accomplishments: This has been a successful IPM project. Gray mold storage decay caused by Botrytis cinerea is the most important postharvest disease of kiwifruit and is commonly responsible for large losses during long-term storage of kiwifruit. Gray mold decay does not occur in the field but is a direct result of B. cinerea infections that do occur in the field but remain latent in the senescent floral parts (mainly sepals) and stem end scars (receptacles).
In this project, it was possible to successfully predict gray mold decay of kiwifruit in storage from the incidence of colonization of sepals and stem ends by B. cinerea. Sampling fruit from vineyards 4 months after fruit set and recording the incidence of Botrytis colonization in sepals or stem ends was used as a field-monitoring method to predict the incidence of kiwifruit Botrytis gray mold after 3 or 5 months in cold storage.
Spraying the fungicide vinclozolin 1 and/or 2 weeks before harvest significantly reduced postharvest gray mold after 3 months storage. Preharvest sprays reduced gray mold only by a little after 3 months storage, but had a major effect when the incidence of gray mold of unsprayed vines was more than 4% in 1995.
In 1995/96, the incidence of colonization of sepals predicted correctly gray mold in five of nine (56%) vineyards and the incidence of colonization of receptacles (stem ends) in seven out of nine (78%) vineyards. The method can be used successfully by growers to make decisions about preharvest sprays, sorting, repacking, and when to market and ship fruit. About two to three growers in 1995 and a dozen of growers in 1996 used the method before making a decision on preharvest spray(s).
DOWNY MILDEW WARNING SYSTEM
Implementation of a disease warning system for downy mildew of lettuce using high resolution weather forecasts and a geographical information system.
Principal Investigator: A. H. C. van Bruggen, Plant Pathology, Davis
Summary of Accomplishments: Currently, growers rely on three to five calendar-based applications of protective fungicides (maneb and/or aliette) for control of lettuce downy mildew (Bremia lactucae). Although sprays are applied frequently, disease pressure is sometimes low and most fungicide applications to a given lettuce crop may be wasted. The overall objective of this research was to develop a warning system for improved chemical control of lettuce downy mildew based on weather forecasts and regional availability of inoculum. The use of forecast weather rather than measured (past) weather could provide sufficient lead time for growers to perform spray applications before infection takes place, thus improving the efficacy of protective fungicides.
Our previous studies in commercial lettuce fields in coastal California had shown that infection with downy mildew occurs mainly during periods of prolonged morning leaf wetness duration (LWD), suggesting that fungicide applications could be timed according to morning LWD. A simulation model for LWD caused by dew was used in conjunction with medium-resolution weather forecasts to predict prolonged morning LWD periods (and thus disease risk) with a lead time of 24 hours. Predicted temperatures, relative humidities, and leaf wetness were compared with observed values. Temperature and relative humidity predictions were generally in agreement with predicted values, except in the morning and evening. The number of hours with leaf wetness was underestimated by weather forecasting companies and our model predictions (except in 1996).
In field trials in 1995, too many mornings with prolonged leaf wetness (including those caused by drizzle) were missed so that disease severity was higher when sprays were applied according to the forecasting system than when sprays were applied on a calendar-based schedule. In 1996, observed and forecast relative humidity and temperature were used to predict survival of spores in the afternoon and infection in the subsequent night, in addition to infection during prolonged morning leaf wetness periods. One morning with prolonged leaf wetness was predicted but did not occur, resulting in one superfluous spray (but savings compared to a calendar-based schedule).
To regionalize the disease warning system, results of field surveys on downy mildew incidence and severity were entered and analyzed in Geographical Information System (GIS). Temperature data from CIMIS stations in the Salinas area were entered into GIS, in combination with a digital elevation map of the region. This part of the project is more complex than anticipated but we got a good start in the development of regional disease warning systems based on local infection risk, observed and predicted weather, and regional disease severity. The development of a regional forecasting system is continued with funding from the USDA NRI program.
TRACKING LEAFMINERS IN LETTUCE
Management of Liriomyza leafminer outbreaks in lettuce.
Principal Investigators: J. Granett, Entomology, Davis; W. Chaney, UC Cooperative Extension, Monterey County
Summary of Accomplishments: We sampled populations of leafminer flies (Liriomyza huidobrensis), in lettuce fields and associated trap catch counts with the sources of those populations. Sources of flies in a newly planted lettuce field are (1) adjacent crops if those crops are good hosts; (2) the previous host-crop in the field; and (3) remote host-crops which flies can be blown from or fly off of. Our data indicate that all three sources operate.
Sources 1 and 2 can be decreased if growers manipulate crop juxtapositions in time and space. That is, a grower might consider leaving a field fallow for the 2 weeks it takes for soil pupae to emerge as adults before replanting and refrain from planting a new lettuce crop in a field adjacent to a host-crop field that is nearing harvest. Source flies can be found in lettuce, celery, cauliflower, broccoli, spinach, green onion, and mustard, with lettuce being the greatest producer of leafminers. We found no evidence that the flies overwinter as pupae in fallow fields, hence fields held fallow for the few weeks it takes for adult eclosion is sufficient to limit this source of flies in a newly planted field. However, since flies also come from remote fields (source 3), such manipulations alone will not be satisfactorily effective. Hence, understanding mortality is important.
Mortality studies in lettuce in 1995 and in lettuce and celery in 1996 showed that flies had almost no parasitism; only 4% parasitism occurred in the celery and 0% in lettuce. In 1995 mortality was high in the egg stage indicating that host factors were causative. Since high egg mortality was not seen in 1996, we surmise that the 1995 environmental conditions were critical. In 1996 we saw a high pupal mortality indicating that this might be a weak link in the insect's life cycle.
These studies indicate that manipulation of abiotic factors might profitably be tested as a way of increasing mortality. In particular we noted that many pupae became infected with soil-borne fungi when held in moist conditions. In addition, we might attribute the egg mortality to concomitant changes in plant turgor or temperature. Although it is culturally inappropriate to grow lettuce under nonoptimal conditions, we suggest tests with brief dry or wet conditions, especially early in the plant's phenology, as a means of changing turgor in plants without impacting plant growth. During the middle part of plant growth, having conditions wetter than normal may serve to increase entomopathogenic fungal activity against pupae. Such mortality may be experimentally increased by fungal treatments.
EVALUATING STRATEGIES FOR MANAGING AN INSECT-TRANSMITTED PLANT VIRUS
The incidence of tomato infectious chlorosis virus in greenhouse and field tomatoes and management strategies for its control.
Principal Investigators: B. W. Falk, Plant Pathology, Davis; S. Temple, Agronomy and Range Science, Davis; R. M. Davis, Plant Pathology, Davis
Summary of Accomplishments: We developed and applied indirect methods to detect the whitefly-transmitted tomato infectious chlorosis virus (TICV) in plants. To accurately detect TICV, we used complementary DNA (cDNA) probes generated by us previously, as well as some new probes constructed by us during this work. A nonradioactive assay was used to positively and accurately detect TICV in fresh or frozen leaf extracts, and results could be collected and found that fresh, frozen, or even air-dried leaves were, in most cases, suitable for accurate TICV detection.
A survey for the incidence of TICV was done by collecting tomatoes from several areas during the late 1995 and 1996 growing seasons, both from greenhouse and field-grown plants. These samples were analyzed for TICV. In 1995 a total of 396 samples were collected from Yolo, San Joaquin, Stanislaus, and Monterey counties.
In 1995 only 57 tomato plants were found to be infected by TICV. Most were older greenhouse-grown plants, but included samples from Yolo and San Joaquin counties. The only TICV-infected field tomatoes were old plants that originated as transplants. These were originally grown in a greenhouse that also contained older TICV-infected tomato plants being grown for seed production. In 1996 we only sampled field tomatoes. TICV was not found in commercial fields, but was recovered from one sample of direct-seeded tomatoes at the UC Davis research farm. High populations of the greenhouse whitefly were also present in this field. Because of the emergence of whiteflies and the threat of new whitefly-transmitted viruses, we developed probes that proved to be effective for detecting three additional whitefly-borne closteroviruses.
IMPROVING NEMATODE SAMPLING STRATEGIES
Sequential decision methods to control sugarbeet cyst nematode via rotations.
Principal Investigators: B. B. Westerdahl, Nematology, Davis; O. R. Burt, Agricultural Economics, Davis; E. P. Caswell-Chen, Nematology, Davis
Summary of Accomplishments: The sugarbeet cyst nematode, Heterodera schachtii, is a widespread, serious pathogen of sugarbeets in California. The damage threshold for beets grown in southern California has been established as one egg/g soil. Because soil temperatures during part of the growing season in northern California are cooler than in southern California, it had been proposed that the damage threshold might be higher in northern California.
To assess the damage threshold in northern California, beet fields scheduled for standard fall harvest and beet fields that were overwintered for spring harvest were used as experimental sites. Research was initiated in a total of 22 fields, with data collected successfully from 18 fields. For each field investigated individual plots were established within the fields. For each plot the following data were recorded: the initial nematode density (eggs per cm3 of soil), the yield, (calculated as tons per acre), and the final nematode density (eggs per cm3 of soil).
Our results clearly revealed that the damage threshold for cyst nematode on sugarbeets in northern California is low, one egg/ cm3 of soil. This damage threshold approaches the detection limit of our sampling techniques. Most surprisingly this damage threshold is appropriate for beets harvested in fall, and for beets that are held over the winter for harvest in spring. Thus, we have determined that the general damage threshold for cyst nematode on beets in California is approximately one egg/cm3 of soil. This result leads us to emphasize the importance of adequate monitoring, rotations, and weed management to reduce damage caused by cyst nematode.
MOLECULAR MARKERS FOR MONITORING LETTUCE DOWNY MILDEW
Development and use of molecular markers for the control of lettuce downy mildew.
Principal Investigators: R. Michelmore, Vegetable Crops, Davis; K. Subbarao, Plant Pathology, Davis; S. Koike, UC Cooperative Extension, Monterey County; F. Laemmlen, UC Cooperative Extension, Santa Barbara County
Summary of Accomplishments: We collected isolates of lettuce downy mildew on a monthly basis throughout the Salinas and Santa Maria growing areas during the 1995 growing season and on an opportunistic basis during 1996 as well as receiving isolates from cooperating pest control advisers. Molecular marker analysis of DNA from single lesions proved more difficult than anticipated. We have continued to refine the techniques and are now close to practical analysis of large numbers of isolates. Due to reduced funding and the need to provide answers on variation in the pathogen, we focused on analyzing isolates using classical techniques. Although fewer isolates were characterized than would have been possible using molecular markers and the data was generated more slowly than ideal, we were able to determine the range and geographical distribution of pathogen variation.
We characterized a total of 374 isolates collected in 1995 for sensitivity to the fungicide Ridomil 2E and a subset for their ability to overcome resistance genes as well as their mating type. Forty-one 1996 isolates were characterized for all three traits. Approximately 20% were sensitive to Ridomil, 20% had intermediate sensitivity, and the remainder were insensitive. Isolates varied widely in their ability to overcome resistance genes. Previously characterized pathotypes were present but did not predominate. A new group of isolates was widespread and was therefore designated Pathotype V. Our data indicates that the pathogen population is in transition from one that is predominantly asexual to a more variable sexual one. Both mating types were detected in commercial fields, although one was rare. None of the rare B1 isolates were Ridomil-insensitive providing the opportunity to slow the transition to a sexual population by using applications of fungicide to epidemics caused by B1 isolates.