Bonanza Creek SiteBytes 2010


This last year at BNZ there has been substantial activity on top of maintaining our core operational systems. Maintaining system integrity and security continues to be our prime task and it requires a considerable amount of resources before we can tackle any larger issues; but there are plenty of additional projects vying for our attention. Preparation and submission of our new research proposal to the National Science Foundation played a leading role for many months. There were two major focus points related to data management addressed in our proposal prepared for NSF. First of all, a great deal of time was spent to update and improve the content and functionality of our website. Secondly, we spent considerable effort to ensure that all core data sets are current and have the most complete metadata.

The website improvements we initiated have significantly improved the quality of our internet presence and outreach capability. While the design of previous web system was somewhat dated and potentially in need of a more complete overhaul we felt limited in the resources we could commit to this effort in light of other ongoing projects. Rather, we chose to upgrade our ColdFusion Web Software and server OS to improve system integrity before enhancing the functionality and content of specific web pages. This provided us with a revitalized platform that could provide more robust services and stability. Content that could be transferred into tables within the database and displayed programmatically was migrated and additional web-based interfaces were created. Additional database content was also displayed on pre-existing and new web pages. Several key researchers contributed additional text and graphics to update and enhance content while new pages were created to highlight recent art and humanities focused projects. While the temptation to migrate to Drupal for site and data management was significant, in the end we really felt that such a conversion was not feasible given the timeframe and the anticipated learning curve. However, this option is still possible for the future as our IM system evolves. As the network continues to move in this direction and offer more support and modules, it will become a more attractive option.

The data review process included reviewing database outputs and EML for core data sets to ascertain their accuracy and completeness. As issues were identified, the appropriate information was gathered and inserted into the database. Procedures for the generation and harvesting of data by network systems were also reviewed and we are currently in the process of updating the scripts for data submissions to ClimDB. Thanks to some improvements with our EML generation scripts, the Metacat harvests are working to pull in the highest quality metadata on record for BNZ.

This past year, Jason Downing also completed serving as editor for the bi-annual IM newsletter LTER Databits. The opportunity to assist in its production for three full issues was educational, and valueable. Jason would like to encourage any and all LTER IM's to volunteer as editor as vacancies emerge and to submit articles whenever possible.

Also, in January of this year, we saw the departure of our information management consultant Rebecca Koskela. Her previous responsibilities have been migrated to our site IM and operations were able to continue without significant interruption. To support this transition and to help enhance the data managers' competence, Jason has been pursuing additional training in database administration, computer programming and scripting languages through a variety of delivery methods.

Our site GIS systems are maintained by Jamie Hollingsworth who has recently been active in the LTERmaps development team. He has helped with development, organization and presentation for a recent ASM session, numerous VTC's and an upcoming workshop to be held later this fall. In conjunction with these efforts, we have reviewed and improved our sites' GIS and spatial data management system and hope to remain a testing ground for future network systems.

Additionally, we have been working to refine and expand the capabilities of our sensor network database and web visualization package. Additional climate data is going through the final stages of a significant review process before it will be used to repopulate this database and be available through the VistaDataVision interface. Some additional core data not previously in a database format has been migrated into this system as well when it seemed appropriate and the visualization capabilities are desirable. As this system is capable of hosting many more sensor stations the BNZ manages, we have been working with other research teams as well as the Arctic LTER to tap into these services and allow them to take advantage of this BNZ hosted system.

Discussion Questions Response:

Do we give a lot of attention to a few datasets or limited attention to a lot?
There is definitely a subset of data sets that gets a more significant level of attention than others. For these data, we actively reach out to core scientists and technicians to make sure that these select data are as current, complete, and accurate as possible because we have determined them to be of the highest value for their duration and versatility. These are also some of the most requested and longest data sets. The next level of priority most likely goes to new dataset additions and from students and researchers. Getting the most current data sets documented and input into our system before personnel changes or data entropy for the given project becomes excessive. Transient researchers provide a valuable data stream to be captured but the window of opportunity is short; when they offer new submissions it is critical to free up the appropriate level of attention before the window closes. Our final level of attention goes out to system-wide metadata and EML improvements that should benefit the entirety of our data catalog. This mostly deals with database design and EML generation procedures.

Do you deliver derived or raw data?
We do both. Core climate data and the like are generally left in its rawest form possible but certain data must be aggregated or processed to some level before it can be utilized. Individual project and student data is generally of a more aggregated type and we generally leave the decision of data aggregation up to the submitting researcher.

Do you communicate with your PIs regarding prioritization of data?
We have recently embarked on a campaign to personalize our core data sets by assigning a senior researcher to have oversight over each of the core data sets. The goal of this effort is to get an individual person to assume an ownership role over each of the long-term data sets that are not directly part of any individual research project but are more communal in nature. The "Core Data Contact" will help to address quality control and methodology issues that are best lead by an individual who has experience with that particular data type.

Information management issues are also discussed at our almost-monthly BNZ-LTER Staff Meetings. These include all the site and data management personnel, the BNZ PI and co_PI’s, as well as any other specific reseachers that may be particularly suited to a particular discussion item. Agenda items cover a variety of topics but data management issues are often the major topic of discussion.