We present a database containing morphometric information about the global distribution of Martian valley networks. The classification of valley networks is based on morphometry only (e.g. width to length relationship, amount of tributaries). This approach avoids possible misinterpretations which might be induced by a genetic classification e.g., sapping valleys). Valleys were mapped on a Viking derived base map (i.e., the MDIM2 of the U.S. Geol. Survey). Elevations were measured in a spatially coregistered, MOLA-based digital elevation model (DEM) with a horizontal resolution of 500 m/pixel. Valleys and tributaries were subdivided into one or several straight segments. For each segment, we measured (a) the length, (b) the azimuth, (c) the elevation difference along the segment, and (d) the geographic coordinates of the start and end points. Four major types of valley networks were classified: 1. longitudinal valley networks have short (i.e. shorter than the width of the main valley) or very few tributaries, 2. dendritic valley networks have more than three tributaries and a width which is comparable to the length of the main branch, 3. radial or (sub)-parallel channels which have no tributaries ("gullies"), 4. valley networks, perpendicular to large canyon and graben systems ("slope valleys"). The compiled web-based database enables us to select all combinations of parameters within a class. A data plot will be generated automatically. The data can be exported to GIS software for postprocessing and analysis. Ongoing work focuses on the evaluation of volumetric (depth, cross-section, volume) and planimetric measurements (width at different locations) to derive flow parameters for each valley network. Morphometric parameters of mapped features on Viking MDIM will be completely reviewed using the MOC-WA global orthoimage map, recently developed at the DLR.