1

I have some general questions about how to go about solving this problem:

I am trying to incorporate FEWSnet's monthly Rainfall Estimate data from 2014-2015 for Southern Africa into an analysis where I am using a combination of survey data that I collected myself (projected in long/lat coords, WGS84 ellipsoid), the FEWSnet data, and some other rasters/SPDF's that already have projections defined (these are not my concern - I have been able to use projectRaster() and spTransform() to coerce these to new coordinate systems as needed).

The FEWSnet data are challenging because according to their documentation (http://earlywarning.usgs.gov/fews/product/120) they are published in an "Albers Equal Area Conic projection that uses the Clarke 1866 spheroid". But, when I download and read the files into R and call crs() on them, I get "CRS Arguments: NA" in return.

Still being a newbie to these things (and being largely self-taught) I'm confused about exactly what information to use to inform how to correctly set the projection of these data. I've tried looking for an EPSG code for AEA projections in Africa, as well as simply searching the internet for a solution, but am coming up empty. Here is the code that I used to import the data:

# Assume that data are downloaded from the above link, and stored in
# a directory on the desktop entitled "Map_Layers". Set this as wd.
# "Map_Layers", in my case, contains data from various sources including FEWS
# and within FEWS subdirectory are further directories for the raster files
# representing year, month, and dekadal periods of rainfall data.

DecRFE1 <- raster("./FEWS/2014_Dec_dd1_RFE/s14121rf.bil")
DecRFE1 # printed output below...
#    class       : RasterLayer 
#    dimensions  : 530, 640, 339200  (nrow, ncol, ncell)
#    resolution  : 8000, 8000  (x, y)
#    extent      : -1456000, 3664000, -4224000, 16000  (xmin, xmax, ymin, ymax)
#    coord. ref. : NA 
#    data source : /Users/HenryHWells92/Desktop/Map_Layers/FEWS/2014_Dec_dd1_RFE/s14121rf.bil 
#    names       : s14121rf 
#    values      : 0, 255  (min, max)

list.files("./Map_Layers/FEWS/2014_Dec_dd1_RFE")
# [1] "s14121rf.bil" "s14121rf.blw" "s14121rf.clr" "s14121rf.hdr"

So, consider this question a general plea for help on several things:

  1. How to choose the correct CRS given the information in the files themselves and in the documentation on FEWSnet's website.
  2. What are the uses of each of the four file types that the FEWSnet data downloads give you? (The .bil files are not in a format that I'm familiar with...)
  3. If the best way forward is to set a custom projection, what information do I need to do so? Is there an EPSG crs I could take advantage of?

closed as too broad by If you do not know- just GIS, Mapperz Jan 19 '16 at 18:52

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • epsg.io/?q=africa+albers only gives one answer and it uses the wrong elipsoid – Ian Turton Jan 19 '16 at 17:27
  • going to cpc.ncep.noaa.gov/products/GIS/GIS_DATA/rfe gives some shapefiles too - which contain GEOGCS["GCS_WGS_1984_Perfect_Sphere",DATUM["D_Sphere",SPHEROID["Sphere",6371000.0,0.0]],PRIMEM["Greenwich",0.0],UNIT["Degree",0.0174532925199433]] but I don't think that helps either – Ian Turton Jan 19 '16 at 17:34
  • 1
    try one question per post, you will have more success. – If you do not know- just GIS Jan 19 '16 at 17:58
  • The reason the reported CRS is NA is that the hdr (header) file of the bil format does not normally store this information. So you need to know it. The 'bil' format is simple, but archaic. – Robert Hijmans Jan 19 '16 at 22:40
  • I thought there was something odd about the .bil format. Good to know my suspicions were along the right lines. – Henry Hawkins Wells Jan 20 '16 at 1:41
3

I downloaded your data and had a play around in QGIS - if I choose EPSG:102022 (Africa_Albers_EAC) it looks like it lines up. Depending on your plans for this that might be close enough - obviously for real science(tm) you'll need to go back to NOAA (or the publications listed) and check. Or you could try editing the spheroid definition below.

enter image description here

PROJCS["Africa_Albers_Equal_Area_Conic",
GEOGCS["GCS_WGS_1984",
    DATUM["WGS_1984",
        SPHEROID["WGS_1984",6378137,298.257223563]],
    PRIMEM["Greenwich",0],
    UNIT["Degree",0.017453292519943295]],
PROJECTION["Albers_Conic_Equal_Area"],
PARAMETER["False_Easting",0],
PARAMETER["False_Northing",0],
PARAMETER["longitude_of_center",25],
PARAMETER["Standard_Parallel_1",20],
PARAMETER["Standard_Parallel_2",-23],
PARAMETER["latitude_of_center",0],
UNIT["Meter",1],
AUTHORITY["EPSG","102022"]]
  • 2
    In R this EPSG code does not work. Is it an ESRI code (see spatialreference.org/ref/esri/africa-albers-equal-area-conic). This suggests that in R you would use +proj=aea +lat_1=20 +lat_2=-23 +lat_0=0 +lon_0=25 +x_0=0 +y_0=0 +ellps=WGS84 +datum=WGS84 +units=m +no_defs – Robert Hijmans Jan 19 '16 at 22:45
  • According to the FEWS documentation, this would be closer: +proj=aea +lat_1=-19.0 +lat_2=21.0 +lat_0=1.0 +lon_0=20 +x_0=0 +y_0=0 +ellps=clrk66 +units=m +no_defs – mkennedy Jan 19 '16 at 23:10
  • Thanks you guys. This is a big help. Regardless of the difficulty of the different formats / software it's good to see how the information all translates. If it doesn't work, I'll update the question. Cheers. – Henry Hawkins Wells Jan 20 '16 at 1:41

Not the answer you're looking for? Browse other questions tagged or ask your own question.