Modeling potential of halophytes in the production of biofuel and edible oil using linear regression and an adaptive neuro-fuzzy inference system

Halophyte plants are capable of tolerating high salinity due to their unique morphological characteristics, vegetative forms, physiological attributes, and specific salt tolerance mechanisms. These plants offer significant economic benefits, prompting research into their potential use in biofuel and edible oil production. This study focused on three plant species: Haloxylon persicum, Tamarix ramossisima, and Halocnemum strobilaceum. Samples were collected at distances of 1000, 1500, 2000, and 5000 m from the Miduk mining site in Shahrbabek, with four replications of 20 individual plants each. Data regarding plant characteristics, seed oil content, ethanol, and lignin production potential were measured and analyzed. The results indicated that Haloxylon persicum has a higher cellulose-to-hemicellulose ratio compared to lignin, making it a suitable raw material for bioethanol production. Additionally, gas chromatography analysis of the oil revealed that halophyte seeds contain ten fatty acids: five saturated fatty acids-lauric acid, myristic acid, palmitic acid, stearic acid, and arachidic acid-and five unsaturated fatty acids-behenic acid, oleic acid, linoleic acid, alpha-linoleic acid, and stearidonic acid. Haloxylon persicum also proved to be more suitable for edible oil production due to its higher oil content. Furthermore, results from the modeling section revealed that the adaptive neuro-fuzzy inference system exhibited greater accuracy than linear regression in predicting biofuel and edible oil parameters in coastal species.