The Live Access Server (LAS) has become a popular system with which to provide Web access to gridded data sets. With a single-window browser interface LAS users can i) specify the location, time and variable of interest , ii) request visualizations along principle axes and planes, and iii) download subsets in a choice of formats. With a few more mouse clicks he can "fuse" (compare by differencing) variables that may be defined on different coordinate grids, stored in unlike file formats, and located at distributed sites. (Other papers will discuss access to distributed data via the Distributed Ocean Data System (DODS)).

The style of access that LAS provides for gridded data is also applicable to large collections of in-situ measurements. Users of measured data often want to view a collection as a whole - a 4-dimensional region in which, although the variables not gridded, they can be subsetted and visualized along lines and planes in much the style of gridded data. A common example arises in collections of ocean profiles where the user may wish to visualize (say) a latitude-longitude plot of temperature at a fixed depth despite the non-uniformity of the data collection techniques.

Recent development work on LAS has focussed on this approach to data management. In this presentation we will demonstrate LAS systems for in-situ collections and discuss the challenges of data fusion for mixed gridded and in-situ data sets. The authors believe that this style of access to in-situ collections will be instrumental in promoting greater use of in-situ measurements in model validation.