The rapid proliferation of the Internet of Things (IoT) technology has paved the way for a more sustainable and efficient approach to environmental monitoring. In this research paper, a comprehensive comparison-based review of energy-efficient strategies in IoT-based machine learning for smart environmental monitoring is presented. This study addresses the need for optimizing energy consumption while maintaining the accuracy and reliability of data collection and analysis in smart environmental monitoring systems. Starting by discussing the evolution of IoT-based environmental monitoring and its significance in the context of creating a smarter and more energy-efficient society. The paper examines a wide range of machine-learning techniques and algorithms commonly employed for environmental data analysis and their respective energy consumption profiles. This findings reveal that, by carefully selecting and implementing energy-efficient machine learning algorithms, it is possible to significantly reduce the power consumption of IoT devices, extending their operational lifespans and reducing the need for frequent battery replacements or recharging. Additionally, the discussion is done on the impact of data compression and transfer techniques, as well as low-power communication protocols, in further enhancing energy efficiency within IoT systems. Also delve into various case studies and practical implementations, highlighting successful approaches and their outcomes in terms of energy efficiency and data quality. The paper concludes by emphasizing the critical importance of continuous research and development in the realm of energy-efficient IoT-based machine learning for smart environmental monitoring, to achieve a more sustainable and eco-friendly smart society.