SUMMARYDue to the potential impact of climate change and climate variability on rainfed production systems, both farmers and policy makers will have to rely more on short- and long-term yield projections. The goal of this study was to develop a procedure for calibrating the Cropping System Model (CSM)-CROPGRO-Soybean model for six cultivars, to determine the potential impact of climate change on rainfed soybean for five locations in Georgia, USA, and to provide recommendations for potential adaptation strategies for soybean production in Georgia and other south-eastern states. The Genotype Coefficient Calculator (GENCALC) software package was applied for calibration of the soybean cultivar coefficients using variety trial data. The root mean square error (RMSE) between observed and simulated grain yield ranged from 201 to 413 kg/ha for the six cultivars. Generally, the future climate scenarios showed an increase in temperature which caused a decrease in the number of days to maturity for all varieties and for all locations. This will benefit late-planted soybean production slightly, while the increase in precipitation and carbon dioxide (CO2) concentration will result in a yield increase. This was the highest for Calhoun and Williamson and ranged from 31 to 49% for the climate change projections for 2050. However, a large reduction in precipitation caused a decrease in yield for Midville, especially based on the climate scenarios of the Global Climate Models (GCMs) Commonwealth Scientific and Industrial Research Organisation's model CSIRO-Mk3.0 and Geophysical Fluid Dynamics Laboratory's model GFDL-CM2.1. Overall, Calhoun, Williamson, Plains and Tifton will probably be more suitable for rainfed soybean production over the next 40 years than Midville. Farmers might shift to a later planting date, around 5 June, for the locations that were evaluated in the present study to avoid potential heat and drought stress during the summer months. The cultivars AG6702, AGS758RR and S80-P2 could be selected for rainfed soybean production since they had the highest rainfed yields among the six cultivars. In general, the present study showed that there are crop management options for soybean production in Georgia and the south-eastern USA that are adapted for the potential projected climate change conditions.

Predicted air and dew point temperatures can be valuable in decision making in many areas including protecting crops from damage, avoiding heat stress on animals and humans, and in planning related to energy management. Current web-based artificial neural network (ANN) models on the Automated Environment Monitoring Network (AEMN) in Georgia predict hourly air and dew point temperature for twelve prediction horizons, using 24 models. The observed air temperature may approach the observed dew point temperature, but never goes below it. Current web based ANN models have prediction errors which, when the air and dew point temperatures are close, may cause air temperature to be predicted below the dew point temperature. Herein this error is referred to as a prediction anomaly. The goal of this research was to improve the prediction accuracy of existing air and dew point temperature ANN models by combining the two weather variables into a single ANN model for each prediction horizon. The objectives of this study were to reduce the mean absolute error (MAE) of prediction and to reduce the number of prediction anomalies. The combined models produced a reduction in the air temperature MAE for ten of twelve prediction horizons with an average reduction in MAE of 1.93 %. The combined models produced a reduction in the dew point temperature MAE for only six of twelve prediction horizons with essentially no average decrease in MAE. However, the combined models showed a marked reduction in prediction anomalies for all twelve prediction horizons with an average reduction of 34.1 %. The reduction in prediction anomalies ranged from 4.6 % at the one-hour horizon to 60.5 % at the eleven-hour horizon.

The accurate prediction of air temperature is important in many areas of decision-making including agricultural management, transportation and energy management. Previous research has focused on the development of artificial neural network (ANN) models to predict air temperature from one to twelve hours in advance. The inputs to these models included a constant duration of prior data with a fixed resolution for all environmental variables for all prediction horizons. The overall goal of this research was to develop more accurate ANN models that could predict air temperature for each prediction horizon. The specific objective was to determine if the ANN model accuracy could be improved by applying a genetic algorithm (GA) for each prediction horizon to determine the preferred duration and resolution of input prior data for each environmental variable. The ANN models created based on this GA based approach provided smaller errors than the models created based on the existing constant duration and fixed data resolution approach for all twelve prediction horizons. Except for a few cases, the GA generally included a longer duration for prior air temperature data and shorter durations for other environmental variables. The mean absolute errors (MAEs) for the evaluation input patterns of the one-, four-, eight-, and twelve-hour prediction models that were based on this GA approach were 0.564°C, 1.264°C, 1.766°C and 2.018°C, respectively. These MAEs were improvements of 3.98%, 4.59%, 2.55% and 1.70% compared to the models that were created based on the existing approach for the same corresponding prediction horizons. Thus, the GA based approach to determine the duration and resolution of prior input data resulted in more accurate ANN models than the existing ones for air temperature prediction. Future work could examine the effects of various GA and fitness evaluation parameters that were part of the approach used in this study.

OBJECTIVE:To evaluate the accuracy of artificial neural networks (ANNs) for use in predicting subjective diagnostic scores of lameness with variables determined from ground reaction force (GRF) data.ANIMALS:21 adult mixed-breed dogs.PROCEDURES:The left cranial cruciate ligament of each dog was transected to induce osteoarthritis of the stifle joint as part of another study. Lameness scores were assigned and GRF data were collected 2 times before and 5 times after ligament transection. Inputs and the output for each ANN were GRF variables and a lameness score, respectively. The ANNs were developed by use of data from 14 dogs and evaluated by use of data for the remaining 7 dogs (ie, dogs not used in model development).RESULTS:ANN models developed with 2 preferred input variables had an overall accuracy ranging from 96% to 99% for 2 data configurations (data configuration 1 contained patterns or observations for 7 dogs, whereas data configuration 2 contained patterns or observations for 7 other dogs). When additional variables were added to the models, the highest overall accuracy ranged from 97% to 100%.CONCLUSIONS AND CLINICAL RELEVANCE:ANNs provided a method for processing GRF data of dogs to accurately predict subjective diagnostic scores of lameness. Processing of GRF data via ANNs could result in a more precise evaluation of surgical and pharmacological intervention by detecting subtle lameness that could have been missed by visual analysis of GRF curves.

Frost damage is responsible for more economic losses than any other weather related phenomenon in the United States (USA) and many other regions across the globe. With sufficient warning, producers can minimize the potential damages caused by frost and freeze events. However, the severity of these events is dependent upon several factors including air temperature, dew point temperature, and wind speed. Methods for assessing this risk are not easily quantifiable and require the insight of experts familiar with the process. Georgia's Extreme-weather Neural-network Informed Expert (GENIE) incorporates the knowledge of expert agrometeorologists and additional information on air temperature, dew point temperature, and wind speed into a fuzzy expert system for use by Georgia producers to provide warning levels of frost and freeze for blueberries and peaches. Artificial neural network (ANN) predictions of air temperature and dew point temperature across the state of Georgia for one to 12 h ahead and observed wind speed are used as input variables for this fuzzy expert system. Meteorological conditions were classified into five levels of frost and freeze by the expert agrometeorologists. These expertly classified scenarios were then used to develop fuzzy logic rules and membership functions for GENIE. Additional scenarios were presented to GENIE for evaluation and it classified all scenarios correctly. This tool will be made available to Georgia producers through a web-based interface, which can be found at www.georgiaweather.net.

Sudden changes in weather, in particular extreme temperatures, can result in increased energy expenditures, depleted agricultural resources, and even loss of life. However, these ill effects can be reduced with accurate air temperature predictions that provide adequate advance warning. Support vector regression (SVR) was applied to meteorological data collected across the state of Georgia in order to produce short-term air temperature predictions. A method was proposed for reducing the number of training patterns of massively large data sets that does not require lengthy pre-processing of the data. This method was demonstrated on two large data sets: one containing 300,000 cold-weather training patterns collected during the winter months and one containing 1.25 million training patterns collected throughout the year. These patterns were used to produce predictions from 1 to 12 h ahead. The mean absolute error (MAE) for the evaluation set of winter-only patterns ranged from 0.514°C for the 1-h prediction horizon to 2.303°C for the 12-h prediction horizon. For the evaluation set of year-round patterns, the MAE ranged from 0.513°C for the 1-h prediction horizon to 1.922°C for the 12-h prediction horizon. These results were competitive with previously developed artificial neural network (ANN) models that were trained on the full data sets. For the winter-only evaluation data, the SVR models were slightly more accurate than the ANN models for all twelve of the prediction horizons. For the year-round evaluation data, the SVR models were slightly more accurate than the ANN models for three of the twelve prediction horizons.

To determine the feasibility of using x-ray technology for internal defect removal, a discrete-event simulation model was developed with the capability to vary inspection machine configuration and placement and associated economic impact. To evaluate the profitability of X-ray inspection, the results of the simulation model were also used to estimate the unit cost (UC) and selling price per box of the inspected onions. A fundamental assumption of this work is that the increase in quality resulting from the X-ray inspection would command a higher market value as reflected by sales price. An Arena simulation model was developed based on data from two packinghouse time studies, from 2005 to 2006 onion X-ray inspection and sample measurements (4 cultivars each year) and from published cost estimates of sweet onion production and packing. The model results indicated the feasibility of incorporating three and four X-ray inspection units at 0.25 m s(-1) belt speed under the simulated conditions to obtain realistic throughput rates. Estimated costs per 18.14 kg box of X-ray inspected onions ranged from US $9.00 to US $15.00 while the estimated selling price ranged from US $11.35 to US $25.34, depending on farm yield, the quality of incoming crop and the gross profit margin goals. The methods could be readily extended to other commodities destined for the fresh market; however, the results would be commodity and site specific. (C) 2009 IAgrE. Published by Elsevier Ltd. All rights reserved.

During the past decade, there have been significant improvements in remote sensing technologies, which have provided high-resolution data at shorter time intervals. Considerable effort has been directed towards developing new classification strategies for analysing this imagery, but the use of artificial intelligence-based analysis techniques has been somewhat limited. The aim of this study was to develop an artificial neural network (ANN)-based technique for the classification of multispectral aerial images for land use in agricultural and environmental applications. The specific land-use classes included water, forest, and several types of agricultural fields. Multispectral images at a 1-m resolution were obtained for the state of Georgia, USA from a Geographic Information Systems (GIS) data clearinghouse. These false-colour images contained green, red and infrared true-colour information. Three approaches were used for the preparation of the inputs to the ANN. These included histograms of the pixel intensities, textural parameters extracted from the image, and matrices of the pixels for spatial information. A probabilistic neural network was used. Seven hundred images were used for model development and 175 for independent model evaluation. The overall accuracy for the evaluation data set was 74% for the histogram approach, 71% for the spatial approach and 89% for the textural approach. The evaluation of ANNs based on various combinations of all three approaches did not show an improvement in accuracy. We also found that some approaches could be used selectively for certain classes. For example, the textural approach worked best for forest classes. For future studies, edge detection prior to classification, with more careful selection of each class, should be included for land-use classification of multispectral images.

Crops and livestock in most of the southeastern United States are susceptible to potential losses due to extreme cold and heat. However, given suitable warning, agricultural and horticultural producers can mitigate the damage of extreme temperature events. To provide such a warning, air temperature prediction models are needed at horizons ranging from 1 to 12h. The goal of this project was to explore the application of artificial neural networks (ANNs) for the prediction of air temperature during the entire year based on near real-time data. Ward-style ANNs were developed using detailed weather data collected by the Georgia Automated Environmental Monitoring Network (AEMN). The ANNs were able to provide predictions throughout the year, with a mean absolute error (MAE) of the year-round models that was less during the winter months than the MAE of the models resulting from the application of previously developed winter-specific models. The prediction MAE for a year-round evaluation set ranged from 0.516°C at the one-hour horizon to 1.873°C at the twelve-hour horizon. A detailed graphical analysis of MAE by time-of-year and time-of-day was also performed. A tendency to over-predict temperatures during summer afternoons was associated with localized cloud cover during that period. The inclusion of rainfall as input to the model was also shown to improve prediction accuracy. In addition, two simple ensemble techniques were explored and neither parallel nor series aggregation was found to reduce prediction errors. When simulated over two extreme temperature events, the models were capable of rapidly adjusting predictions on the basis of new information. The final models were applied to prediction horizons of 1–12h and deployed on the website of the Georgia AEMN (www.GeorgiaWeather.net) for use as a general, year-round decision support tool.

Lack of information on postharvest packinghouse performance hinders exploration, assessment of improvement opportunities and education possibilities. This study evaluated the sizing and inspection performance of 3 onion packinghouses and developed a discrete event simulation model to demonstrate the impact of improving these 2 performance variables on potential sales revenue generation, as part of a larger goal to develop a methodology for bringing packinghouses into the teaching and demonstration classroom via simulation. A group of 550 fresh sweet onions from the Vidalia production region in Georgia, US were obtained from 3 packinghouses for the 2-performance variable evaluation. Results indicated significant difference (p < 0.05) among the 3 packinghouses in terms of sizing error rate. The major departure from homogeneity was caused by a relatively higher fraction of incorrectly sized onions in 1 packinghouse. There was no significant difference (p > 0.05) between the packinghouses in terms of percentage rejects in the sorted Grade 1 onions. One packinghouse failed to meet the tolerance limit for defects, as specified by the US Grade Standards. Packinghouse managers were polled to discern impact, with houses responding with major management and packing line modifications. (C) 2008 IAgrE. Published by Elsevier Ltd. All rights reserved.

Dew point temperature is needed as an input to calculate various meteorological variables. In general, it contributes to human and animal comfort levels. The goal of this study was to develop artificial neural network (ANN) models for dew point temperature prediction to improve upon previous research. These improvements included optimizing the stopping criteria, comparing seasonal models to year-round models, and developing ensemble ANNs to blend the output of seasonal models. For an ANN trained with 100,000 patterns per epoch, the error was reduced using a 2000-pattern stopping dataset at an interval of 20 learning events to decide when to stop training. Seasonal ANN models were blended in an ensemble ANN with the weight of the member networks determined using a fuzzy membership-type function based on the day of year. These ensemble models were shown to produce lower errors than year-round, nonensemble models. The mean absolute errors (MAEs) of the final models evaluated with an independent evaluation dataset included 0.795°C for a 2-hour prediction, 1.485°C for a 6-hour prediction, and 2.146°C for a 12-hour prediction. The final model MAEs, when compared to the previous research, were reduced by 0.008°C, 0.081°C, and 0.135°C, respectively. It can be concluded that the methods used in this research were effective in more accurately predicting year-round dew point temperature. The ANN models for different prediction periods were sequenced to provide a 12-hour dew point temperature prediction system for implementation on the Georgia Automated Environmental Monitoring Network website (www.georgiaweather.net).

Dewpoint temperature, the temperature at which water vapor in the air will condense into liquid, can be useful in estimating frost, fog, snow, dew, evapotranspiration, and other meteorological variables. The goal of this study was to use artificial neural networks (ANNs) to predict dewpoint temperature from 1 to 12 h ahead using prior weather data as inputs. This study explores using three-layer backpropagation ANNs and weather data combined for three years from 20 locations in Georgia, United States, to develop general models for dewpoint temperature prediction anywhere within Georgia. Specific objectives included the selection of the important weather-related inputs, the setting of ANN parameters, and the selection of the duration of prior input data. An iterative search found that, in addition to dewpoint temperature, important weather-related ANN inputs included relative humidity, solar radiation, air temperature, wind speed, and vapor pressure. Experiments also showed that the best models included 60 nodes in the ANN hidden layer, a +/- 0.15 initial range for the ANN weights, a 0.35 ANN learning rate, and a duration of prior weather-related data used as inputs ranging from 6 to 30 h based on the lead time. The evaluation of the final models with weather data from 20 separate locations and for a different year showed that the 1-, 4-, 8-, and 12- h predictions had mean absolute errors (MAEs) of 0.550 degrees, 1.234 degrees, 1.799 degrees, and 2.280 degrees C, respectively. These final models predicted dewpoint temperature adequately using previously unseen weather data, including difficult freeze and heat stress extremes. These predictions are useful for decisions in agriculture because dewpoint temperature along with air temperature affects the intensity of freezes and heat waves, which can damage crops, equipment, and structures and can cause injury or death to animals and humans.

The mitigation of crop loss due to damaging freezes requires accurate air temperature prediction models. Previous work established that the Ward-style artificial neural network (ANN) is a suitable tool for developing such models. The current research focused on developing ANN models with reduced average prediction error by increasing the number of distinct observations used in training, adding additional input terms that describe the date of an observation, increasing the duration of prior weather data included in each observation, and reexamining the number of hidden nodes used in the network. Models were created to predict air temperature at hourly intervals from one to 12 hours ahead. Each ANN model, consisting of a network architecture and set of associated parameters, was evaluated by instantiating and training 30 networks and calculating the mean absolute error (MAE) of the resulting networks for some set of input patterns. The inclusion of seasonal input terms, up to 24 hours of prior weather information, and a larger number of processing nodes were some of the improvements that reduced average prediction error compared to previous research across all horizons. For example, the four-hour MAE of 1.40°C was 0.20°C, or 12.5%, less than the previous model. Prediction MAEs eight and 12 hours ahead improved by 0.17°C and 0.16°C, respectively, improvements of 7.4% and 5.9% over the existing model at these horizons. Networks instantiating the same model but with different initial random weights often led to different prediction errors. These results strongly suggest that ANN model developers should consider instantiating and training multiple networks with different initial weights to establish preferred model parameters. Keywords—Decision support systems, frost protection, fruit, time-series prediction, weather modeling

Artificial neural networks (ANNs) were developed to predict air temperature in I h increments from I to 12 h in the future. Weather data for model development and evaluation for three locations in Georgia (Fort Valley, Blairsville, and Alma) were obtained from the Georgia Automated Environmental Monitoring Network (AEMN). The data consisted of observations of meteorological variables such as air temperature, relative humidity, wind speed, rainfall, and solar radiation. The critical inputs for each model were determined by developing ANNs that used them in various input combinations and observing their effect on the accuracy of the ANN predictions. The results showed that of the meteorological variables considered, only rainfall was not useful in generating air temperature predictions. The optimal duration of prior data ranged from 2 h to 6 h, depending on the period of prediction. The mean absolute error (MAE) increased as the period of prediction got longer The MAE of the evaluation dataset for predicting temperature 1 h in advance was 0.6 degrees C for Fort Valley, 0.7 degrees C for Blairsville, and 0.6 degrees C for Alma. The corresponding MAE values for a 12 h prediction were 2.4 degrees C, 3.0 degrees C, and 2.6 degrees C. Further efforts will be directed to developing general ANNs based on data from multiple locations. The availability of decision support systems that incorporate localized temperature predictions for use by fruit growers could have a positive impact on frost damage protection.

Aflatoxin contamination in crops of peanuts is a problem of significant health and financial importance, so it would be useful to develop techniques to predict the levels prior to harvest. Backpropagation neural networks have been used in the past to model problems of this type, however development of networks poses the complex problem of setting values for architectural features and backpropagation parameters. Genetic algorithms have been used in prior efforts to locate parameters for backpropagation neural networks. This paper describes the development of a genetic algorithm/backpropagation neural network hybrid (GA/BPN) in which a genetic algorithm is used to find architectures and backpropagation parameter values simultaneously for a backpropagation neural network that predicts aflatoxin contamination levels in peanuts based on environmental data.

Finding synchronizing sequences for the finite automata is a very important problem in many practical applications (part orienters in industry, reset problem in biocomputing theory, network issues etc). Problem of finding the shortest synchronizing sequence is NP-hard, so polynomial algorithms probably can work only as heuristic ones. In this paper we propose two versions of polynomial algorithms which work better than well-known Eppstein's Greedy and Cycle algorithms. Keywords—Synchronizing words, reset sequences, Cerný Conjecture.

A systematic approach is developed for modeling permitivity measurement probes. The technique was used to model a commercially available probe. This model was functionally equivalent to the provided model developed with traditional electromagnetic techniques. The approach utilizes the results of a dimensional analysis and a genetic algorithm to find an appropriate model for a probe. The only information required is uncorrected reflection-coefficient measurements on a variety of materials of known permittivity. The need to perform an electromagnetic analysis based on the geometry of a given probe and its construction is not required. This allows candidate probes that cannot easily be analyzed by traditional techniques to be considered for use.

Crop growth simulation models use weather data such as temperature, solar radiation, and rainfall to simulate crop development and yield. The crop models are often needed for locations with missing or incomplete observed weather data. An accurate estimation of these weather variables has thus become necessary. Artificial neural network (ANN) models can be used to accurately interpolate these weather variables, based on neighboring weather stations. The goal of this study was to develop artificial neural network models for estimating and interpolating daily maximum air temperature, minimum air temperature, and total solar radiation for Tifton (south Georgia) and Griffin (north Georgia). Historical daily weather data from 1996 to 1998 were used for model development, and data from 1999 to 2000 were used for final model evaluation. The results of the study indicated that the preferred input variables were straight line distance (∆s) and the elevation difference (∆z) between the target location and input weather stations as well as the values of the variable being estimated at the input stations. Maximum temperature was also found to improve the accurate in estimating solar radiation. The optimum number of input weather stations for estimating each weather variable for both target locations was also determined. The best models for estimating these weather variables were compared with other spatial interpolation techniques, including inverse distance, average, and multi-linear regression methods. The results showed that ANN and regression models provided superior accuracy over inverse distance and average methods. ANN models and regression models were comparable in estimating maximum temperature. The ANN model was clearly more accurate than

APPROACH by RAMYAA (Under the Direction of Ronald W. McClendon) ABSTRACT Air temperatures below freezing can damage plants. Irrigation is the most widely practiced frost protection measure. However, growers need information about when to start irrigating, as the process has to be commenced prior to the temperature dropping below freezing. The goal of this study was to develop Artificial Neural Networks (ANNs) to predict if frost would occur during the near future. A classification approach to develop the ANNs was used. This would require a method to predict frosts, but a model for frost prediction would typically require access to local weather. Many locations that could potentially benefit from frost prediction do not have historical weather data, or even a local weather station. An additional goal was to develop ANNs to predict frost for any given location in the state of Georgia. ANNs were developed using weather data from multiple locations and were evaluated for other locations.