BP-ANN for Fitting the Temperature-Germination Model and Its Application in Predicting Sowing Time and Region for Bermudagrass

Abstract

Temperature is one of the most significant environmental factors that affects germination of grass seeds. Reliable prediction of the optimal temperature for seed germination is crucial for determining the suitable regions and favorable sowing timing for turf grass cultivation. In this study, a back-propagation-artificial-neural-network-aided dual quintic equation (BP-ANN-QE) model was developed to improve the prediction of the optimal temperature for seed germination. This BP-ANN-QE model was used to determine optimal sowing times and suitable regions for three Cynodon dactylon cultivars (C. dactylon, ‘Savannah’ and ‘Princess VII’). Prediction of the optimal temperature for these seeds was based on comprehensive germination tests using 36 day/night (high/low) temperature regimes (both ranging from 5/5 to 40/40°C with 5°C increments). Seed germination data from these temperature regimes were used to construct temperature-germination correlation models for estimating germination percentage with confidence intervals. Our tests revealed that the optimal high/low temperature regimes required for all the three bermudagrass cultivars are 30/5, 30/10, 35/5, 35/10, 35/15, 35/20, 40/15 and 40/20°C; constant temperatures ranging from 5 to 40°C inhibited the germination of all three cultivars. While comparing different simulating methods, including DQEM, Bisquare ANN-QE, and BP-ANN-QE in establishing temperature based germination percentage rules, we found that the R2 values of germination prediction function could be significantly improved from about 0.6940–0.8177 (DQEM approach) to 0.9439–0.9813 (BP-ANN-QE). These results indicated that our BP-ANN-QE model has better performance than the rests of the compared models. Furthermore, data of the national temperature grids generated from monthly-average temperature for 25 years were fit into these functions and we were able to map the germination percentage of these C. dactylon cultivars in the national scale of China, and suggested the optimum sowing regions and times for them.

Application of Genetic Algorithm to Predict Optimal Sowing Region and Timing for Kentucky Bluegrass in China

“…The performances of GA-BP-ANN temperature-germination response models generated in this study were compared with the previously published regression approaches including general quadratic and BP-ANN based quintic equations [ 10 , 12 ].…”

“…These previous models were mainly used to predict: I. the time, under a constant temperature condition (cumulative temperature), required for the expected germination of a specific variety [ 4 – 6 ], and II. the germination percentage under a temperature fluctuation regime [ 10 – 12 ]. It is well-documented that the natural fluctuations in temperature between day and night could be required for initiating and/or facilitating seed germination [ 8 ], these diurnal fluctuations of temperature are frequently adopted to generate data for building prediction models [ 13 ].…”

“…It is well-documented that the natural fluctuations in temperature between day and night could be required for initiating and/or facilitating seed germination [ 8 ], these diurnal fluctuations of temperature are frequently adopted to generate data for building prediction models [ 13 ]. To date, the core functions of the published temperature-germination response models [ 11 , 12 ] consist on the estimation of populations’ thermal response parameters [ 14 – 18 ] or on optimization of polynomial equations using iterative curve fitting [ 6 , 10 ]. These functions constructed by various scientists are usually different from each other because their parameters are selected for fitting the germination data of a particular batch of seeds.…”

“…In an effort to simplify and standardize the selection of core function for a temperature-germination model, we propose a GA-based data mining approach that automatically generates a core function of seed germination and temperature correlation. Meanwhile, the back propagation (BP) algorithm is also recruited to optimize the GA based temperature-germination model [ 10 ].…”

An Artificial Neural Networks‐Based Tree Ring Width Proxy System Model for Paleoclimate Data Assimilation

“…Additionally, when using ANN model to solve real-world problems, clear prior knowledge of the input-output rules behind the real-world problem to be solved is not needed (Basheer & Hajmeer, 2000;Micheli-Tzanakou, 2011). Thus, ANN model has already been used for building "black-box" models (Lantz, 2015) in a wide range of fields, including the geosciences (Altunkaynak, 2007;Fang et al, 2012;Haq & Menon, 2014;Pham et al, 2017;Pi et al, 2013;Woodhouse, 1999). Generally, a black-box model refers to a model or system in which both inputs and outputs can be observed, but the mechanism by which inputs are mapped to the outputs is not observed and, typically, not fully understood.…”