Aim :Methodology :
Results :Interpretation :Approaches to modelling pest populations range from simple empirical models to advanced soft computing techniques that have advantages as well as limitations.A comparative analyses of modelling approaches result in selection of betterpest forecast model with a higher prediction accuracy.
Artificial neural network (ANN) techniques, multi-layer perceptron neural network (MLP-NN) and polynomial neural networks (PNN) were used along with the multiple and polynomial regressions to predict the moth population of tobacco caterpillar (Fabricius) in groundnut cropping system. pheromone trap catch and weather data of twenty five years (1990-2014) for season (26 to 44 standard meteorological weeks (SMW)) was used for predictive modelling. The weekly male moth catches of (numbers/trap/week) during maximum severity period (34 SMW) was modelled using weather variables maximum and minimum temperature (°C), rainfall (mm), morning and evening relative humidity (%) lagged by two weeks. The performance of the models was evaluated using coefficient of determination (R ), root mean square error (RMSE) and mean absolute percentage error (MAPE) estimates.The study clearly demonstrated the superiority of MLP-NN (R :0.89) over all other models for predicting the peak severity of . Sensitivity analysis of MLP-NN model indicated that the maximum temperature lagged by two weeks and evening relative humidity of the previous week was two most important factors influencing the peak population of . Validation also demonstrated the effectiveness of MLP-NN followed by PNN in dealing with non-linear relation between population and weather variables.All model equations developed in the present study can be used to predict peak (34 SMW) in conjunction with weather of 32 and 33 SMW during season, and in issuing need based advisories for its effective management on groundnut.
A multi-layer perceptron (MLP) neural network model for predicting adult moth population of tobacco caterpillar (Spodoptera litura (Fabricius) in groundnut cropping system of Dharwad (Karnataka) was developed and tested using the long term (24 years : 1990-2013) trap catches of the pest and weather data of Kharif season [26 to 44 standard meteorological weeks (SMW)]. The weekly male moth catches of S. litura during maximum severity observed at 34 SMW was modelled using the weather parameters viz., maximum temperature (C), minimum temperature (°C), rainfall (mm) and morning and afternoon relative humidity (%) lagged by two weeks. The principle component analysis was performed using meteorological data of preceding two weeks (32 and 33 SMW) in order to create fewer linearly independent factors. Five principal component scores which together accounted for 90 per cent of variations in data were used as input variables for neural network model. A MLP neural network with five input nodes and one hidden layer consisting of eleven hidden nodes was found to be suitable in terms of adequacy measures for modelling the population dynamics of S. litura. While data sets of 1990-2009 were used for developing the model, data of four seasons (2010-2013) were used for testing and validation. The performance of the model was assessed in terms of root mean square error (RMSE) and mean absolute percentage error (MAPE). The validation results clearly showed that the neural network based model is effective in dealing with the apparently random behaviour of the S. litura dynamics on groundnut.
Modeling of seed sorption kinetics have been routinely applied for predicting the equilibrium moisture content (EMC ) of processed products and the information is used for determining their drying and storage parameters. Similar information is also valid for the seed industry because moisture content of the stored seed would determine its longevity and quality. But in seed industry, the storage facilities are used for multiple species having distinct physical and biochemical characteristics and there is a need to explore and identify a single mathematical model that can comprehensively predict moisture sorption behaviour of a wide range of species. In the present study, the Modified Henderson model could effectively describe the sorption kinetics of Pea, Okra and Chilli and gave the best fit for the data, when analyzed using mean relative error (MRE, %), standard error of estimate/moisture (SEM) and randomness of residual pattern as observed in the residual plots.
In this paper, an ordinal logistic regression model was developed for predicting the severity of tobacco caterpillar, Spodoptera litura (Fabricius) on groundnut using the pest dynamics vis a vis climatic data of twenty five years (1990-2014) pertaining to Kharif (26 to 44 standard meteorological weeks (SMW)) season of Dharwad (Karnataka). Trend analysis of climatic data using Mann-Kendall nonparametric test showed that mean and minimum temperatures, and rainfall to be increasing while morning and evening relative humidity and their mean to be decreasing over time. The weekly male moth catches of S. litura (nos./trap/week) during maximum severity period (34 SMW) was modeled with climatic variables lagged by two weeks. The developed model indicated that the maximum temperature and morning relative humidity prior to two weeks contributed significantly to the occurrence of high level of pest attack. Results suggested that for each degree increase in maximum temperature during 32 SMW, the odds of being high pest attack (as opposed to lower or medium) increased by a multiple of 8.6 as compared to the odds of being high or medium (as opposed to low) increasing by 6.4 times for each per cent rise in the morning relative humidity.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.