Hi, I am struggling with this quite a while already, and I haven't found a good solution so far. I want to disaggregate precipitation data. For this I have a grid with daily precipitation in 1x1 km resolution, and additionally 24 grids with 10x10 km resolution giving percentage charge of hourly precipitation for the same day as factors (range 0 to 1). The first grid is in UTM coordinates and the other 24 are in longlat coordinates. Now, I want to transfer the factors from the 10x10 km grids to grids of the same size as the 1x1 km daily precipitation grid. What I have done so far is: 1. Generate SpatialPoints with the longlat coordinates for the 10x10 km grid (comming as netCDF). 2. Convert longlat coordinates to UTM coordinates (spTransform), adding the data values to a SpatialPointDataFrame 3. Interpolate data from the converted SpatialPpointDataFrame to a grid of the same size and extention as my 1x1 km grid by using the function interp() from the akima package. However, instead of interpolating I'd rather like to use a kind of nearest neigbour classification because I want to avoid "smoothing" of my factors. Furthermore, interp() is relatively time consuming. Since I have to repeat this procedure for many days, it would be nice to have a quicker alternative. Does anyone have an idea? Any hint is much appreciated. Klaus -- View this message in context: http://r-sig-geo.2731867.n2.nabble.com/downscaling-grid-data-alternative-for-interp-tp6819843p6819843.html Sent from the R-sig-geo mailing list archive at Nabble.com.
downscaling grid data; alternative for interp()?
2 messages · Klaus Vormoor, Paul Hiemstra
Hi, A trick to get Nearest Neighbor (NN) is to use idw() from the gstat package with setting the nmax parameter to 1. This limits the amount of points used for the inverse distance weighted average to one, leading to NN. I'm not sure if this is much faster than interp, but you can take a look. regards, Paul
On 09/22/2011 11:45 AM, Klaus Vormoor wrote:
Hi, I am struggling with this quite a while already, and I haven't found a good solution so far. I want to disaggregate precipitation data. For this I have a grid with daily precipitation in 1x1 km resolution, and additionally 24 grids with 10x10 km resolution giving percentage charge of hourly precipitation for the same day as factors (range 0 to 1). The first grid is in UTM coordinates and the other 24 are in longlat coordinates. Now, I want to transfer the factors from the 10x10 km grids to grids of the same size as the 1x1 km daily precipitation grid. What I have done so far is: 1. Generate SpatialPoints with the longlat coordinates for the 10x10 km grid (comming as netCDF). 2. Convert longlat coordinates to UTM coordinates (spTransform), adding the data values to a SpatialPointDataFrame 3. Interpolate data from the converted SpatialPpointDataFrame to a grid of the same size and extention as my 1x1 km grid by using the function interp() from the akima package. However, instead of interpolating I'd rather like to use a kind of nearest neigbour classification because I want to avoid "smoothing" of my factors. Furthermore, interp() is relatively time consuming. Since I have to repeat this procedure for many days, it would be nice to have a quicker alternative. Does anyone have an idea? Any hint is much appreciated. Klaus -- View this message in context: http://r-sig-geo.2731867.n2.nabble.com/downscaling-grid-data-alternative-for-interp-tp6819843p6819843.html Sent from the R-sig-geo mailing list archive at Nabble.com.
_______________________________________________ R-sig-Geo mailing list R-sig-Geo at r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Paul Hiemstra, Ph.D. Global Climate Division Royal Netherlands Meteorological Institute (KNMI) Wilhelminalaan 10 | 3732 GK | De Bilt | Kamer B 3.39 P.O. Box 201 | 3730 AE | De Bilt tel: +31 30 2206 494 http://intamap.geo.uu.nl/~paul http://nl.linkedin.com/pub/paul-hiemstra/20/30b/770