Issue 15708

Analytics: Generate the ES50 index

15708
Reporter: trobertson
Type: Feedback
Summary: Analytics: Generate the ES50 index
Priority: Minor
Status: Open
Created: 2014-05-28 09:30:12.423
Updated: 2014-05-28 09:30:12.423
        
Description: A request to do some data analysis from Mark Costello

Mark Costello:
"Have you considered coding ES50 index into GBIF for (say 5-degree cells) to produce an ‘annual’ global richness map for example? Perhaps even run it for time periods of data like <1900, - 1920 -1940 – 1950 -60-70-80-90-2000-2010 to see if the pattern changes? For comparison actual records and species counts would indicate sample bias which is also interesting."

"Thinking of the GBIC in july I wondered whether to try to do something by then. Is it possible to get the number of records and number of ‘species’ (ideally those matched to WoRMS and CoL but perhaps any name that passes some basic tests of seeming a real name) for every 1 or 5-degree lat-long cell? This may be the first land&sea map of world biodiversity based on actual data. Of course species will be biased by sampling effort (records) but I think the ES50 index is the best way of trying to account for this. Perhaps you know it? It scales 0-50 and calculates the number of species in 50 randomly chosen records for each ‘sample’ (perhaps a 5-degree cell as there may be too few species in many 1-degree cell). There is free software online to calculate it.

I have a PhD student brilliant with GIS and maybe we can plot species per altitude and depth along some transects (e.g. latitude) as graphs – should show  striking richness in tropics and coastal zones if you guys are stuck for time."

]]>