Lots of incorrectly georeferenced records in Herbarium of Taiwan Forestry Research Institute dataset

18321
Reporter: rdmpage
Type: Feedback
Summary: Lots of incorrectly georeferenced records in Herbarium of Taiwan Forestry Research Institute dataset
Priority: Unassessed
Status: Open
Created: 2016-03-14 15:57:07.452
Updated: 2016-03-14 16:28:41.206
        
Description: Looking at the Herbarium of Taiwan Forestry Research Institute dataset http://www.gbif.org/dataset/98b2bb30-1627-11df-bd84-b8a03c50a862 there are points in strange places, such as the sea or Asian plants in Africa. I suspect this dataset has some serious georeferencing issues.

For example, the forest tree _Podocarpus macrophyllus_ http://www.gbif.org/occurrence/237163261 is given as Shihtzu Hsiang(獅子鄉) but the coordinate splice it in the sea off Sulawesi. I suspect the latitude should be 22.17369 not 2.17369. Likewise, the occurrence http://www.gbif.org/occurrence/237075291 of Broussonetia papyrifera in Egypt seems unlikely, the locality is given as 株洲郊區樟樹下 which translates as "Camphor tree Zhuzhou suburb" and the original record http://taif.tfri.gov.tw/search_spec.php?SpcmID=110335&DupID=1&SeriID=1&DetID=222180&l=Cht gives no indication this specimen comes from Egypt.

It would be nice if the providers checked all records outside Taiwan to see how many are correct.]]>
    

Author: mblissett
Created: 2016-03-14 16:27:01.37
Updated: 2016-03-14 16:28:41.202
        
There's a grid pattern over northern Africa.  That pattern could be a way to spot errors like this.  I wonder if it's a format string like ##.###### that has caused many records to be 1/10th around the world from where they should be.

The dataset was last successfully crawled in 2013, and the Tapir installation doesn't seem to be working.