|
The US National Park Service (NPS) preserves the natural night sky, to the greatest extent possible, as a natural resource and value. Our first step towards understanding and protecting the night sky is to assess the night sky quality. In the parks, we capture a series of overlapping high-resolution images to obtain a mosaic view of the entire night sky. The NPS Night Skies Program has specialized pipeline to perform data reduction and processing of these 45 one-million-pixel images taken by our mobile camera system. Specifically, the pipeline applies basic reduction, performs photometric calibration, mosaics the images to form the panoramic views of the entire night sky, generates the models for the natural sky brightness, and calculates values of different sky brightness metrics. This original data pipeline was developed mainly by a single person about a decade ago. These scripts are written in three different languages including Python, Java, and Visual Basic and interact with six different proprietary software packages including ACP Observatory Control, ArcGIS, MaxIm DL, Microsoft Excel, Photoshop, and Pinpoint. Although the current data pipeline is complete and functional, an upgrade is needed to make the pipeline distributable and manageable. Recognizing the challenges of moving forward, NPS Night Skies Program recently formed a pipeline development team for this upgrade effort. We, the members in the pipeline development team, are a multidisciplinary group of NPS staff with background in astronomy, geography, biology, and engineering. Geographically, we are located in both the mainland US and Alaska. Our goal is to transform the entire pipeline to be completely written in Python and minimize the number of required proprietary software. In addition, we will allow any future pipeline updates or new development to be made or suggested by any user instead of from a single-person initiated top-down model. For our pipeline upgrade project, we are using Git and GitHub for version control, project management, and source-code distribution. Once the upgrade is completed, the pipeline will be distributed via GitHub to NPS staff and partners located in Alaska, Nevada, Colorado, Tennessee, and Texas. All of our Python pipeline scripts will be open source where we hope to also benefit other scientists in the similar research field worldwide.
|
