Dear List,
I have not heard back, so my apologies if someone has already suggested
something better, but I have found a solution and am replying to myself for
others who may have this problem in the future.
If, for my sparse data, I comment out and do not use the line:
gridded(sparse.frame) <- TRUE
Then the spatial analysis works out quite fine.
Thank you,
Cora
===================================
Date: Thu, 23 Oct 2008 19:09:33 -0600
From: "Cora Shea" <[EMAIL PROTECTED]>
Subject: [R-sig-Geo] Mapping sparse data evenly to a
SpatialPixelsDataFrame
To: r-sig-geo@stat.math.ethz.ch
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain
Hello everyone,
Thank you in advance for any help; my apologies for the basic nature of my
question.
Here goes...I have three vectors, one with my data, and the other two with
the integer x coordinates and integer y coordinates of the data. Put
together theoretically, they form a sparse sampling of a 250x250 point
grid. For example, a sample would be a data value of "1100" located at
"121,34" where 1100, 121, and 34 are the i-th values in each of the three
vectors.
For comparison, I have a fully filled 250x250 grid (62500 data points), in
the same three-vector format.
When I make a SpatialPixelsDataFrame object with the -full- 62500 data
points, eg
> full.frame <- data.frame(z = full.data, x.coord, y.coord)
> coordinates(full.frame) <- c("x.coord", "y.coord")
> gridded(full.frame) <- TRUE
.....then summary(full.frame) gives:
cellcentre.offset cellsize cells.dim
x.coord 1 1 250
y.coord 1 1 250
...as desired. Yay! But when I do the same thing building a data.frame
with the three 'sparse' data vectors (all three vectors have length of 129,
which is the number of samples, which is << 62500):
> sparse.frame <- data.frame(z = sparse.data, x.sparse.coord,
y.sparse.coord)
> coordinates(sparse.frame) <- c("x.sparse.coord", "y.sparse.coord")
> gridded(sparse.frame) <- TRUE
.....summary(sparse.frame) gives:
cellcentre.offset cellsize cells.dim
x.sparse.coord 0 3.387097 75
y.sparse.coord 2 3.240000 77
Which is not desired, they should be 1,1,and 250 as above, right?
Additionally, when forming sparse.frame I get the following:
Warning messages:
1: grid has empty column/rows in dimension 1 in: points2grid(points,
tolerance)
2: grid has empty column/rows in dimension 2 in: points2grid(points,
tolerance)
I would like to be able to perform various functions (eg a variogram) on the
sparse samples, but such functionality requires square cells.
So, after extensively trying to use the SpatialGridDataFrame class instead
of the SpatialPixelsDataFrame one and bbox and so on, I'm coming up short
with a way to do something like:
> gridparameters(sample.frame)$cellsize[1] <- 1
Which doesn't work.
In short, how can I force the sparse samples to be a given unit square
width?
Thank you very much for your time,
Cora
[[alternative HTML version deleted]]
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
