GOAL

In the past 25 years SEP sponsors have provided SEP with numerous datasets. Early datasets were a few cmps or shot gathers. Over time SEP has added large 2-D surveys, multi-component dataset, 3-D prestack and post-stack, and even 4-D time lapse datasets. Unfortunately these datasets were not carefully cataloged and or generally left in easily accessible state. As a result, finding a dataset to apply a particular technique to could be an arduous task.

The goal of the SEP datalib Mini-Seminar was to make finding and using SEP's datasets easier. Each dataset was researched and processed so the following form could be filled out:

• Location of raw data
• Location of velocity model
• Stack of data
• Zero offset Migration
• Past usage of data
• Geometry of data
• Characteristics with data
• History of data (when acquired and by whom)
• Preprocessing on data
• Proprietary considerations

The inexpensive cost of large disks made it possible to put a large portion of the data online. All 2-D datasets and portions of every 3-D dataset are now stored online. The maturity of SEP3d Biondi et al. (1996); Clapp (1999) allowed us to maintain a full description of each dataset while allowing the dataset to be sliced and diced.