PSICAT in Antarctica

When I finally deployed to Antarctica to support PSICAT, I did so with some trepidation. I expected that there would be the inevitable bugs to fix and new features to implement. I was reasonably certain that PSICAT would not be a complete failure, but it was a little nerve-wracking to have a $30 million project depending on my unproven software. During my three months in Antarctica I was stationed in McMurdo where I shared cramped sleeping and living quarters with the other members of the ANDRILL team and worked 14-hour shifts, seven days a week. Living and working in such close quarters with the users meant that there was no place to hide when things went wrong. This situation provided ample motivation to really listen to the users and keep them happy. After the first few weeks of intense interaction with the users and bug fixing, PSICAT's performance improved to the point that I came to believe that it was really going to work. The habit of frequent releases helped me to address the users' requests quickly. In the three months on the ice, I released 30 updates with bug fixes, feature enhancements, and new features, an average of one update every three days. I was able to address most requests in one or two days. The ones that I couldn't address right away usually involved larger-scale modifications to make existing features easier or more intuitive to use. Since I wanted to undertake large changes after giving proper thought to the design and implications of these changes, I made a note of suggestions for investigation and inclusion in future releases.

In the end, PSICAT's performance exceeded everyone's expectations and the additional benefits of using it contributed significantly to the success of ANDRILL's first drilling expedition. Since PSICAT provided instant access to the core description diagrams, it allowed the core to be presented and discussed on the same day it was described by the sedimentology team. Gone were the days of crude photocopies of hand-drawn description diagrams and waiting a day or two for the initial digital versions to be available. Being able to present the diagrams at the morning meetings allowed the scientific team to discuss the core in detail and develop a sampling strategy for further scientific analysis. Another key benefit of using PSICAT was the ability to summarize the data automatically. In the past, the sedimentology team would spend many hours drafting summary diagrams by hand at the end of the expedition. Now summaries could be generated instantly and on-demand. This allowed the science team to see the emerging trends in the core as the drilling progressed, an enormous advantage for understanding the geological evolution of the core. Finally, I observed that PSICAT enabled better science by allowing the science team to revise the data freely to improve its accuracy. In the past, changing some aspect of the core description half-way through the drilling would have required going back and re-drafting every diagram and was obviously not routinely done. Instead, with PSICAT these changes could be executed in seconds resulting in more accurate diagrams.

About ANDRILL

ANDRILL (ANtarctic geological DRILLing) is a multinational collaboration comprised of more than 200 scientists, students, and educators from five nations (Germany, Italy, New Zealand, the United Kingdom and the United States) to recover stratigraphic records from the Antarctic margin using Cape Roberts Project (CRP) technology. The chief objective is to drill back in time to recover a history of paleoenvironmental changes that will guide our understanding of how fast, how large, and how frequent were glacial and interglacial changes in the Antarctica region. Future scenarios of global warming require guidance and constraint from past history that will reveal potential timing frequency and site of future changes.

Acknowledgments
I would like to take this opportunity to thank everyone involved in making PSICAT a success. First and foremost, I'd like to thank my advisor, Dr. Cinzia Cervato, for finding me such a challenging and interesting problem to work on, and for keeping me in the graduate student lifestyle I've grown accustomed to. On the geology side of things, I'd like to thank Dr. Chris Fielding for his help in explaining the intricacies of the core description process, and on the technology side of things I'd like to thank Doug Fils and Xiaoyun Tang for their insightful guidance in designing PSICAT. And finally, I'd like to say thanks to the ANDRILL project for putting their faith in PSICAT, and a special thank you to the MIS sedimentology team. I can't imagine a better group of people to spend three months in Antarctica with.