rast | Globally optimal geometric matching

I have used many times the same process of rasterization, which works fairly well:

I am using the focal function in the terra package to fill holes in a DEM raster. However, after running focal(m, w=19, fun=mean, na.policy="only", na.rm=TRUE) my DEM is moved ~11cm relative to it's original position (see images below). This problem does not occur when I simply read in the raster using rast() and then write it using writeRaster(), but it happens when I use focal() regardless of whether I save the output using writeRaster() or using the filename = argument within focal().

I have also noticed that the larger the window size used, the greater the movement of the output raster. A window size of 9 produced ~5cm error. Raster pixels are ~1.2cm in size so focal() appears to be causing a movement of ~1/2 the width of the window used. As far as I can tell the movement is straight up/north.

Any idea why this might be occuring and how it could be avoided?

This the correctly aligned raster:

And this is after running focal (w = 19):

I am trying to count the non-NA values in a spatRaster using the global() function from the terra package. All the functions (mean, max, sd, etc.) seem to work except for "isNA" and "notNA". For these two functions it returns this error: Error in fun(values(x[[i]]), ...) : could not find function "fun", which is the same error it returns for a misspelled/non-existent function.

I want to crop a raster using a bbox or a known extent, i.e., 10 pixels in row and col.

Below you can see a reproducible example:

I want to select a group of pixels from a raster.

• The user should enter an n value for selecting the number of columns and rows (or window size).

• A window would fetch the number of pixels defined.

• Would return a raster with the values.

Below is a piece of reproducible example:

PREFACE My error is a simple "Mathematical operation applied to non-numeric argument" error, but I think this arise from how I create a suite of user defined functions and use them within the terra::app() function. I am going to describe the full workflow to elucidate what I am after, so please bare with me.

ISSUE I am trying to apply a statistical-empirical topographic correction to Sentinel 2A data in R. To apply the topographic correction, I have appended rasters of solar azimuth, solar zenith, and slope and aspect to the multiband scene. I am first taking a random sample of each band within a scene to get its intensity value as well as the corresponding solar zenith, aspect, and slope values. From there, I calculate the cosine of the solar incident angle for each sampled cell using the zenith, azimuth, slope and aspect using a user-define function. Then, I run a linear regression between the cosine of solar incident angle and the respective intensity value for each band. I then apply a user-defined function that calls the solar incident function above using the terra::app() function to finalize the topographic information from these linear regressions. This works on one core just fine with fake data, but is painfully slow with real Sentinel data, so I want it to work on multiple cores. When I try to run on multiple cores I get the error:

I want to perform a neighborhood analysis in R to smooth the layer I have but keeping all the NAs of the input raster intact.

However, when I do, for instance, the following, the calculation "propagates" over the NA values - what it is an undesiderable behavior, in my case.

I am very new to R and programming in general, please forgive my forthcoming ineptitude.

I am working with a large categorical raster. Essentially, every pixel shallower than 10 meters on the Great Barrier Reef is assigned a value: 11,12,13, or 15. (Original file here). My end goal is to create a histogram showing the frequency of the "rubble" category (which is the given by the value 12) by latitude. It would look very similar to the third panel of this figure, but where they show "coral habitat" I would be showing rubble.

I thought the best way to do this would be to try to convert the original raster into a data frame where each row represents what was a pixel and there are three columns: categorical value (11,12,13, or 15), latitude, and longitude. I could then use this data frame to create any number of basic plots.

Ideally I would like to omit NAs in the process of creating this data frame because the raster is 152,505 by 112,421, but over 99% of pixels are empty (due to the shape of the Great Barrier Reef).

I can easily read in the raster and plot it using the Raster or Terra packages:

With terra I would like to place the legend horizontally below the map. From help it seems that this can be done by either passing a character to legend (e.g. legend = "bottomleft") or by passing a list to plg with arguments from the base-R legend function. However, for me none of these work. Please, see code sample below.

Thanks!