Greetings, all
I am just notifying of a couple of errors in two of the ERDDAP datasets. I don think there is anything to do other than replace the underlying data, but I could be wrong.
I have emailed the listed contact (
jay.h...@noaa.gov) for this particular dataset but I haven't received a response.
IBTRACS are data on tropical cyclones, with the full dataset dating back to the 1850s.
There are two IBTRACS datasets listed on the tabledap page, both are supposed to be subsets of the full IBTRACS data.
- IBTRACS_last3years
- IBTRACS_since1980
Unfortunately, both of these ERDDAP datasets have underlying filters that constrain their actual representation. Both only contain data from 2022 to 2025. For the "_last3years" subset this is close enough (for now), but its obviously not good for the "_since1980" subset.
Also, both datasets have one crucial missing column, which identifies the quadrant of the storm. While it seems the existing rows correspond to storm quadrants, one for each of the 4 cardinal directions of NE, SE, SW, NW, the column identifying the quadrant is missing. There are 4 entries for each TIME value, with all values in the 4 identical TIME rows being identical except for the usa_r64, usa_r50, and usa_r34 columns (and the non usa analogs, bom_r64, bom_r50, vom_r34 etc.). However, the crucial quadrant indicator seems to have been dropped. The original IBTrACS data, as found
here, contains the USA_R64_NE, USA_R50_NE etc. columns that indicate which quadrant each wind extent belongs to. The ERDDAP data, however, has columns for usa_r64, usa_r50 etc. but no indication of which quadrant. This information is necessary for proper wind speed representation.
I appreciate the maintenance of these data!
Ben