Acid Rain and Seed Germination: A Predictive Model Using ML-based CART Algorithm

Abstract

The impact of acid rain on the germination of seeds is a significant concern in agricultural and environmental studies. Acid rain, characterized by elevated acidity levels due to pollutants like sulfur dioxide and nitrogen oxides, can adversely affect the germination process of various plant species. The objective of this study was to evaluate the impact of simulated acid rain (SAR) on the germination of Brinjal (Solanum melongena Linn.) and Cowpea (Vigna unguiculata ssp. cylindrica L. Walpers) crops. The experiments were conducted using eight plastic trays of approximately 25 cm. x 30 cm dimensions. Four trays were used for experiments with brinjal seeds (Set I), while the other four were used for cowpea seeds (Set II). One tray from each set was used as positive control and treated with normal pH 5.6, while the other three trays from each batch were treated with SAR solutions of pH 4.5, 3.5, and 2.5. Brinjal seed germination percentage and seed vigor were inferior to Cowpea seeds. The seeds treated with SAR (pH 4.5, 3.5, and 2.5) showed hindered seed germination. Furthermore, a more significant inhibitory effect was observed at lower pH values. The mean germination percentage of seeds was highest for standard SAR (pH 5.6) in the case of Brinjal seeds, while it was recorded lowest for Cowpea seeds. The results indicate that plants do not respond uniformly to SAR. To investigate the behavior of the simulated acid rain data, a Machine Learning-based Decision Tree Algorithm was employed to identify and optimize conditions. Cowpea was predicted to get 95% seed germination, whereas brinjal would only be 64% in acid rain of pH value 5.05 for 36 hours. In conclusion, utilizing a Machine Learning-based CART algorithm has provided valuable insights into predicting the germination behavior of seeds under the influence of acid rain.