Skip to content

a spatial data set recommendation, please

3 messages · Hodgess, Erin, Tomislav Hengl, Tim Appelhans

Hodgess, Erin
# 
Hello!

Could anyone recommend some good spatial data sets with one or two variables with 5000+ observations, please?

I've been looking, probably not in the right place. Just getting little sets.

Thanks,
Erin


Erin M. Hodgess
Associate Professor
Department of Mathematics and Statistics
University of Houston - Downtown
mailto: hodgesse at uhd.edu
Tomislav Hengl
# 
Hi Erin,

I think you are looking for a point data set that is available under 
some public access license? For which area: global? regional/national? 
local?

One data set I like to use for geostatistical mapping is the National 
Geochemical Survey database (http://mrdata.usgs.gov/geochem/) because it 
is very consistent, multivariate and well documented. Another large 
point data set under public access is the NCSS Soil Characterization 
Database (http://ncsslabdatamart.sc.egov.usda.gov/), but these are only 
USA-available of course.

For global data sets, I would look at for example at the daily meteo 
data from Global Historical Climatology Network 
(https://www.ncdc.noaa.gov/data-access/land-based-station-data/land-based-datasets/global-historical-climatology-network-ghcn), 
which I think is partially available via the meteo package 
(http://www.rdocumentation.org/packages/meteo/versions/0.1-5).

Does it has to be inside a package? I've put in my GSIF package small 
sample of the geochemical DB: 
http://gsif.r-forge.r-project.org/geochm.html but it contains 'only' 
2681 points :(

HTH
On 29-7-2016 6:00, Hodgess, Erin wrote:
Tim Appelhans
# 
There is ggmap::crime with 80k+ observations. Again, these are point data
On 29.07.2016 06:00, Hodgess, Erin wrote: