A week of software testing and data manipulation (and lots of rain).
We have begun to test the Semaphore workflow tools against a data set that has been converted for use with the Daycent modelling tools. This data is not yet publicly available so we are unable to share the exact details of the process. But what follows are some Kepler screenshots that hopefully present a clearer idea of where the software product is heading.
To get started, the Kepler software is downloaded from https://kepler-project.org/users/downloads.
The sample Semaphore workflow is then downloaded. In this instance it has been designed to work with this specific dataset.
We are dealing with soil organic flux data, rainfall and temperature data in this set. It has been adjusted to work with Daycent and stored at a path which has been set in the sample workflow’s parameters.
When Kepler is opened (as shown above) we can access the Kepler workflow archive to open a new instance of Kepler, with the Semaphore workflow example.
By pressing the play button in the Kepler GUI, (shown above) the model will be executed, running on the data from the configured pathway.
Data management is another important consideration for this project, as our data co-ordinator, Siobhann explains:
The data that will fuel the Semaphore prototype is taken from two, long-term research sites; the Brigalow Catchment and the Hermitage Station. The information that we have from these two sites is a complex collection of weather and soil data, gas flux measurements and land management information, all of which are typically recorded in Excel files. This has been a challenge for us because the data often needs to be repackaged in a format appropriate for storage in our repository and also for processing by the modelling tools. This is also a challenge because of the size of the data files. Some of the data from both sites span 30 years and the Excel rows are in the tens of thousands. We have had to make sure that the repository upload software we use is able to process this volume, to ensure data is uploaded in full.
We will continue to work on data management issues as we strive to meet our remaining project deliverables. Prominent on the project team’s horizon is Deliverable 7, particularly the automated feed of RIF-CS records. More to come on this as work continues…
Semaphore 