Vegetable producers and marketers make business decisions based on supply estimates. The U.S. Dept. of Agriculture provides estimates of planting intentions for field crops but not for most vegetable crops. This study developed models that can be used to forecast vegetable crop plantings. Multiple linear regression analysis was used to determine the factors that influence plantings of potatoes and onions. Field crop planting intentions, industry structure, lagged values of plantings, prices received, price volatility, and the price of sugar beets were found to be significant factors. The models and/or methods used in this study should be useful to those interested in forecasting vegetable plantings.
Joseph F. Guenthner
Maude Lachapelle, Gaétan Bourgeois, Jennifer R. DeEll, Katrine A. Stewart, and Philippe Séguin
observations of vascular browning in ‘McIntosh’ and ‘Cortland’ apples, developed with weather data from five regions in Quebec from 1977 to 1995, and implemented in the Computer Center for Agricultural Pest Forecasting (CIPRA) software ( Bourgeois et al., 2014b
M.H. Maletta, W.P. Cowgill Jr., W. Tietjen, S.A. Johnston, T. Manning, and P. Nitzsche
Five variations of TOM-CAST and two sources of weather data were used to schedule tomato early blight control for research trials at the Snyder Research and Extension Farm, Pittstown, N.J. TOM-CAST scheduled fungicide applications were initiated at 15, 25, or 35 disease severity values (DSV) and resprayed at 15 or 25 DSV. Weather data for generating the DSVs was obtained on-site with a Sensor Instruments Field Monitor™ or through subscription to the electronic meteorological service SkyBit, Inc. Bravo 720, 3 pints/acre, was used for disease control. Foliar disease, yields, and postharvest decays were evaluated. Daily DSVs, cumulative DSVs, and forecast spray schedule varied with weather data source. Because SkyBit data generated more DSVs during the season than Field Monitor data, the SkyBit-based forecasts called for one or two more sprays than the Field Monitor-based forecasts. However, the number of sprays actually applied was the same, one more or one less for each combination of initiation and respray thresholds. All treatment schedules reduced disease compared to the untreated control. Variation in initiation threshold did not affect disease control. All TOM-CAST schedules respraying at 15 to 20 DSV were as effective as the weekly schedule. All fungicide treatments increased total yields and reduced postharvest decays compared to the untreated control. Most treatments also increased marketable yields. The most efficient, effective Field Monitor-generated TOM-CAST schedule required nine sprays compared to 13 weekly sprays. The comparable SkyBit-generated schedule called for 10 applications. Chemical name used: tetrachloroisophtalonitrile (chlorothalonil).
L.W. Lass, R.H. Callihan, and D.O. Everson
Predicting sweet corn (Zea mays var. rugosa Bonaf.) harvest dates based on simple linear regression has failed to provide planting schedules that result in the uniform delivery of raw product to processing plants. Adjusting for the date that the field was at 80% silk in one model improved the forecast accuracy if year, field location, cultivar, soil albedo, herbicide family used, kernel moisture, and planting date were used as independent variables. Among predictive models, forecasting the Julian harvest date had the highest correlation with independent variables (R2 = 0.943) and the lowest coefficient of variation (cv = 1.31%). In a model predicting growing-degree days between planting date and harvest, R2 (coefficient of determination) = 0.85 and cv = 2.79%. In the model predicting sunlight hours between planting and harvest, R2 = 0.88 and cv = 6.41%. Predicting the Julian harvest date using several independent variables was more accurate than other models using a simple linear regression based on growing-degree days when compared to actual harvest time.
William H. Tietjen, Winfred P. Cowgill Jr., Martha H. Maletta, Peter J. Nitzsche, and Stephen A. Johnston
The effect of disease forecasting systems and stake or ground culture on foliar and postharvest disease control for tomato (Lycopersicon esculentum) was evaluated during two growing seasons in northern New Jersey. Foliar disease was reduced and marketable yield increased by stake culture. Percent of postharvest losses, including loss due to anthracnose, was significantly reduced by stake culture. Effectiveness of disease control schedules, weekly or forecaster-generated, was not affected by cultural system. Disease forecasting was shown to have potential for optimizing fungicide use in tomato production by controlling foliar disease and fruit anthracnose with fewer applications than a weekly schedule.
W.P. Cowgill Jr., M.H. Maletta, and S.A. Johnston
Two disease forecasting systems - FAST, Pennsylvania State University and CUFAST, Cornell University - were used to generate spray schedules for controlling Alternaria solani Ell. and Mart. on `Celebrity' tomato (Lycopersicon esculentum Mill.) at The Rutgers Snyder Research and Extension Farm in Northwest New Jersey. Disease control was compared to that obtained following standard weekly spray schedules. Chlorothalonil, 1.5 lb/A, was used for disease control for all treatments. Disease ratings of the FAST and CUFAST plots were significantly lower than that of the unsprayed control and were not significantly different from the plots sprayed according to standard spray schedules. A total of 10 fungicide applications were made following FAST recommendations; 7 applications were made following CUFAST recommendations; 13-15 applications were made following standard recommended schedules. Using CUFAST resulted in an estimated $200 per acre savings in spray costs. Chemical name used: tetrachloroisophtalonitrile (chlorothalonil).
M.H. Maletta, W.P. Cowgill Jr., W. Tietjen, P. Nitzsche, and S.A. Johnston
The number of fungicide applications for tomato early blight control required by three disease forecasting systems—FAST, Pennsylvania State Univ., CUFAST, Cornell Univ., and TOMCAST, Ridgetown College, Ont.—was less than the number required following a weekly schedule. Foliar disease was significantly lower for all schedules compared to the untreated control. Cultural treatment had no significant effect on disease control, but disease incidence was significantly lower for stake culture than ground culture treatments. Total yield was not affected by cultural treatment, was significantly increased by a weekly fungicide application schedule, and was not appreciably different among the forecast fungicide application schedules. Marketable yield was significantly higher for stake culture than ground culture treatments and was significantly increased by all fungicide application schedules compared to the untreated control. Marketable yield was significantly lower for certain forecast schedules compared to the weekly schedule. Potential cost savings of $379 per acre and pesticide reductions of 33 lbs a.i. per acre for the season were calculated. Chemical name used: tetrachloroisophtalonitrile (chlorothalonil).
M.H. Maletta, W.P. Cowgill Jr., T. Manning, W. Tietjen, S.A. Johnston, and P. Nitzsche
Weather information has many applications in crop production practices, including disease forecasting. A variety of weather instruments are available for on-farm use, but associated costs and need for regular calibration and maintenance can limit actual use, especially by smaller growers. Subscription to an electronic meteorological service may be a viable alternative to on-site weather stations. In 1997 and 1998, hourly temperature, relative humidity and leaf wetness were monitored at six sites in a 400-m2 area of New Jersey with Field Monitor™ data loggers (Sensor Instruments, Inc.) and by subscription to SkyBit, Inc., an electronic meteorological service. There was close correspondence in temperature data from the two sources at all sites, the average seasonal difference ranging from 0 to 2 °F. Relative humidity data was variable between the two sources, the greatest variation occurring at low and high humidity, the ranges at which relative humidity sensors had been shown to be least accurate. Leaf wetness estimates from the two sources agreed at least two-thirds of the time. Data differences related to source were attributed to both systematic and random error. The usefulness of electronic weather data in crop production depends on how sensitive the particular weather-dependent applications (e.g., predictive disease and insect models) are to variation in the input data. The TOM-CAST early blight forecaster for tomatoes was not particularly sensitive to differences between SkyBit and Field Monitor leaf wetness estimates.
M.H. Maletta, W.P. Cowgill Jr., W. Tietjen, S.A. Johnston, and P. Nitzsche
Fourteen different fungicide schedules for early blight control, including eight variations of TOM-CAST, were evaluated at the Snyder Research and Extension Farm, Pittstown, N.J. Weather data was collected with Sensor Instruments Field Monitors. All calendar-based schedules—weekly, biweekly, grower simulation—reduced foliar disease compared to the untreated control. All forecast generated schedules—TOM-CAST variations, FAST and CUFAST—reduced foliar disease compared to the untreated control. Several of the forecast schedules resulted in disease ratings not significantly different from those following calendar based schedules or from each other. The fourteen different schedules required as many as sixteen to as few as four fungicide applications. Disease control schedule did not affect total yield, marketable yield and postharvest losses. Disease control with a TOM-CAST generated schedule based on weather data from an electronic meteorological service was not different from disease control obtained with a TOM-CAST schedule based on ground station weather data. Potential cost savings of as much as $295 per acre resulting from reduced fungicide schedules were estimated. Chemical name used: tetrachloroisophtalonitrile (chlorothalonil).
Arlie A. Powell, Karl Harker, Roger Getz, and Eugene H. Simpson
In order to provide timely weather information to county agents (CEA) and growers, a sophisticated user friendly weather information program was developed that provides over 900 weather files daily to users. This program uses a 420 Sun Server that automatically downloads files from the NWS office on the AU campus and makes them instantly available to CEA offices via the Extension Network. Growers may obtain information from CEAS or use their personal computers to access a “Weather Board”. A chilling/growing degree hour (GDH) model (mod. 45) has been developed for peaches that provides a good estimate of when rest is completed and allows prediction of phenological stages through flowering. This information assists growers with orchard management decisions. Studies with peaches were conducted using the chilling/GDH model to properly apply hydrogen cyanamide (Dormex) to replace lack of chilling. This work resulted in an effective application timing based on chilling accumulation and allowed development of a forecast model for grower use.