In the dynamic landscape of cloud computing, orchestrating scientific workflows efficiently presents a significant challenge. This paper introduces a pioneering task clustering approach designed to streamline the execution of scientific workflows in dynamic cloud environments. The primary goal is to optimize task execution while achieving resource consolidation. Utilizing Directed Acyclic Graphs (DAGs) as workflow representations, our approach incorporates advanced algorithm, Dynamic Cloud Task Clustering (DCTC). Its core objective revolves around enhancing parallelism among scientific tasks, alleviating system overhead, and addressing resource inefficiencies. By efficiently identifying and grouping related tasks into clusters, the approach reduces scheduling overhead for fine-grained scientific tasks, fostering improved parallel processing in dynamic cloud environments. This research represents a significant contribution to the optimization of scientific workflow execution within dynamic cloud environments, with the primary aim of enhancing performance and achieving resource consolidation. A series of rigorous experiments consistently validates the effectiveness of our approach, thus offering promising prospects for more efficient and resource-conscious scientific computing practices in the continuously evolving landscape of cloud computing.