Konza Prairie SiteBytes 2006

LTER Site: Konza Prairie LTER

Contributor: Jincheng Gao (Sep 16, 2006)

Site Byte:

With 18 months work experience as an Information Manager at the Konza site, I have learned a lot and have benefited from the efforts of our former Information Managers, including John Briggs, Brent Brock and Zhiqiang Yang. In particular, I appreciate the efforts of Zhiqiang Yang, who did an excellent job in building the KNZ metadata database and establishing the dynamic web site. In the past year, I have continued to work on the transfer of data from ascii text files to a SQL Server database and QA/QC checking, as well as GIS data generation and regular updating of the database and website.

So far, we have finished the transfer of historical data from text files into the SQL Server database. The data can now be searched and downloaded based on specific watershed, treatment, or date of data collection criteria. Data input interfaces for each dataset were created with VC# to convert text data files into the SQL Server database. Data quality checking is performed based on data range and logical consistency before data transfer. The data ranges are defined according to expected minimum and maximum values for each dataset, and logical consistency is checked based on general knowledge for each data type. For example, daily mean temperature must be less than the daily maximum and greater than the daily minimum values, or the data are flagged for further checking and correction. We are continuing to update the EML for each dataset to comply with the Best Practice for the Metacat harvest. Currently, over 50 datasets in the KNZ database (more than 80% of our core datasets) are harvested weekly by the Metacat database. After QA/QC checking, four stream flow datasets from the KNZ are being regularly harvested by HydroDB.

Based on the suggestion of the KNZ site review last year, I have also collected and edited the GPS data for historical and long-term sites of data collection, such as the climate and stream gauging stations, and long-term transects for collection of plant species data. The accuracy of other spatial data in the KNZ database has been checked, and updated where necessary. The GIS and remote sensing data in KNZ have been also reorganized and their metadata have been updated with FGDC and EML format. The dataset of annual burning history of various watersheds at KNZ is now stored in two formats. One format is as layers for the various watersheds or fire treatments, and stored as a polygon feature class in our database. The other format is as non-spatial data. The non-spatial data are linked with KNZ watershed spatial data through ArcSDE. The burning history data can be interactively queried by watershed name, burning type, and burn time (year or date) in the KNZ interactive web site.

A new bison dataset was created last year. A GPS collar unit was installed on an adult bison as an initial trial to assess the utility of this approach for tracking bison movement patterns. The geographic coordinates of the bison are automatically collected every two hours. The GPS data of the bison are in the interactive web site. As more animals are fitted with tracking devices, the GPS data for bison movement will be integrated with the vegetation data in KNZ to study bison behavior and their effects on heterogeneity in the tallgrass prairie landscape.

In the coming year, I will focus on our web design and the QC/QA checking of our EML metadata. In addition, the methodology associated with each dataset will be converted from the current text format to EML metadata.