The data collections provide products via two data streams with varying latency and quality. Further details are provided below.
Near Real-Time (NRT) data stream
These collections aim to provide all hosted datasets in near real-time (NRT), shortly after the products have been produced by Bureau of Meteorology (BoM) systems.
In some instances, this is not possible due to agreements with third parties. In these cases, we aim to provide the datasets after a fixed delay. Products that are provided in near real-time can be found under a nrt directory.
Key points to note about NRT products:
- Only the last three months of a product will be available. Older products will be deleted. Prior to deletion the products will either be made available under the Archive (ARC) data stream or in the case of Himawari Standard Data (HSD) a different project all together (pc59).
- Products under this data stream will not be gap-filled or reprocessed in the event of an outage.
- Products in this data stream may be of a lower quality than their Archive (ARC) equivalents.
For these reasons, it is advised that the Archive (ARC) equivalents of products be used where possible.
Archive (ARC) data stream
Products in the Archive (ARC) data stream have been gap-filled and (re)processed on Gadi, using the highest quality input data available. Processing is typically done monthly, within six to twelve weeks of the NRT products being made available.
Key points to note about ARC products:
- The whole supported time period will be available.
- Gap filling and (re)processing on Gadi occurs to ensure the data is of the highest quality.
For these reasons, it is advised that the Archive (ARC) equivalents of products be used where possible.