The complexity of data management is increasing, posing additional problems to organizations as they need to gain insights from a large amount of information at their disposal in order to make better decisions and add value to their services or products. Data lifecycles models include the phases and activities that data must go through from creation to disposal. There are several benefits of using these models. They are used to reduce complexity of planning and handling data, to assist stakeholders in better understanding of the data available to them, and to comprehend how data is transformed into knowledge. Different data lifecycle models are necessary to process data and to meet different processing requirements and objectives. Despite the benefits of data lifecycle models, the modelling needs to be improved in order to reflect these necessary variations and provide pertinent information to the involved stakeholders, and support them in decision-making. Data lifecycle models are frequently incorporated into enterprise models to allow expressing the alignment of data lifecycles with services and technical infrastructures. A common approach to model enterprises is Enterprise Architecture (EA), which is a conceptual blueprint that presents a holistic view of an organization's business processes and IT assets, as well as their relationships. Data lifecycle models can be applied in several domains, for example research, semantic web, open data, and smart cities. With the increasing complexity of data exchanges, understanding data lifecycles has become increasingly important in the last few years. For instance, this is particularly noticeable in smart cities, which are a dynamic ecosystem that faces many challenges in order to meet the needs of its citizens. These challenges include managing various stakeholders, gathering data from various sources, integrating heterogeneous data, and needing different processing times. According to the literature, smart cities can be viewed as urban enterprises with complex systems that must be integrated across multiple domains. A data lifecycle is relevant to manage business data throughout its lifecycle while taking into account the constraints of business processes and allowing data requirements to be identified. Existing EAs for smart cities do not identify concepts for describing and modelling data lifecycle variations, which are required to process data. This study suggests a metamodel for improving the modelling of data lifecycles. By designing the proposed metamodel, the elements of a data lifecycle, its variations, and design requirements are identified. Furthermore, the concepts and their relationships identified to model variations in data lifecycles are demonstrated and evaluated in two case studies in this thesis. As a result, this study contributes to a better understanding of the data lifecycle variations and their requirements.