Metadata - "data about data" - aim at describing a data resource and it is an important factor for increasing the exploitation and acceptance of geographical data. It forms a basis for catalogue services and helps users find data suitable for a particular application and assess different data sets. The benefits of metadata will be accentuated as the amount and variety of geographical data increase. Existing metadata standards for geographic data are expressive, yet formulated in a quite general manner. As such, they lack explicit formalized support for describing typical characteristics of 3D/elevation data, e.g., that vertical and horizontal uncertainty can differ or that there is a range of different elevation data types (e.g., the elevation of the bare earth, the highest elevation, bathymetric depth, etc). Such information can be added to the metadata through descriptive keywords or through extending the standard with new or modified code lists of allowed code words. Although using pre-defined code lists limits the expressive power, it eliminates the disadvantages associated with the arbitrariness of keywords. This is an important point, as some confusion seems to exist in this field concerning the interpretation of elevation-related terms. Relevant quality-related information in metadata helps users judge the reliability of data and determine for what the data could be used. This is particularly important for new users with limited knowledge and no or little previous experience of similar data. Quality information can also be exploited in (GIS) data analyses, e.g., in order to combine multiple heterogeneous data sets into refined products. By taking the quality of the respective data layers into account (e.g., horizontal accuracy), a more detailed analysis can be performed, improving the possibility to interpret the result of the analysis. Known uncertainties in data can be tracked through the analysis process, in order to give users and commanders a more solid basis for decision-making. In the report, this is demonstrated in a scenario where possible helicopter landing zones are automatically detected by combing different sources of information (elevation data and aerial photographs). This example illustrates the more general problem of handling heterogeneous data sets. In order to determine what information metadata should contain, a structured analysis of user needs and associate requirements can be performed. In the report, shows how these specifications could be produced by system development methods, exemplified by the helicopter landing zones case outlined above.