Re: [R-sig-Geo] Kriging question, please

This is great! Thanks Joe! Sincerely, Erin Erin Hodgess, PhD mailto: erinm.hodg...@gmail.com On Thu, Dec 19, 2019 at 12:47 PM Joe Lewis wrote: > Hi Erin, > > Depending on what you're trying to do, you may want to look into point > pattern analysis, particularly on linear networks (lots of res

Re: [R-sig-Geo] Kriging question, please

Hi Erin, Depending on what you're trying to do, you may want to look into point pattern analysis, particularly on linear networks (lots of research on how to conduct kernel density estimations with traffic accident data) See spatial analysis along networks (Okabe and Sugihara, 2012) and chapter 1

Re: [R-sig-Geo] Kriging question, please

Hi again everyone: Sorry for the unclear question. Typically, I have data that is at say, 8 locations and collected every hour. So spatial and/or spatial-temporal kriging works fine. In this case, I am dealing with traffic accident data. So it can be at "any" location and at any time. I had t

Re: [R-sig-Geo] Kriging question, please

Howdy, Danie Kriege [dɑniː ˈkriχə] was a mining engineer. I find it very hard to believe there is a practicing engineer who thinks that you can find a gold mine only if you have collected data in a regular grid (if that is what you mean with “r

Re: [R-sig-Geo] Kriging question, please

Hi Erin, I'm not sure what you mean by "irregular data." There's no requirement that points be on a grid, and that's the only thing I can think of, but I'm not sure how you'd get the idea that that was required given the existence of the basic tutorials using the meuse dataset. https://rpubs.com/

[R-sig-Geo] Kriging question, please

Hello! Is there a method for Kriging for irregular data, please? I’m about 99.9% sure that there is not, but just thought I would double check. Thanks so much for your help! Happy Holidays! Sincerely, Erin -- Erin Hodgess, PhD mailto: erinm.hodg...@gmail.com [[alternative HTML versi

Re: [R-sig-Geo] Kriging and plotting categorical variables

As an efficient alternative to kriging you could try using RF with buffer distances to classes incorporated into the model: https://peerj.com/articles/5518/ Also described in: https://envirometrix.github.io/PredictiveSoilMapping/soilmapping-using-mla.html#spatial-prediction-of-soil-types New ki

[R-sig-Geo] Kriging and plotting categorical variables

Dear all, I am working with a set of categorical variables labelled: Group1, Group2, Group3, and Group4. I created a new categorical variables where I replaced the observation Goup1 to Group4 by 1 to 4. Then I performed Ordinary Kriging on the new variable [1-4]. But the outcome of the krigin

[R-sig-Geo] Kriging and Spatio temporal kriging on air quality data

Hi, I am using kriging and spatio temporal kriging for studying and analyzing air quality data from mobile sensors. I would like to ask some question about the following issues: 1) is possible to gave the same values scale to different kriging result (maps from different measures in the same study

Re: [R-sig-Geo] kriging with external drift on precipitation data

See page 16: http://people.ku.edu/~gbohling/cpe940/Variograms.pdf The various models only differ in the way they capture the ascent to spatial autocorrelation until the range. Subtle differences—in my experience (for epidemiological and biological data) an exponential model suffices. typing wit

[R-sig-Geo] kriging with external drift on precipitation data

Hi I'm not a geostatistic expert. I'm trying to interpolate raingauges precipitation measures using radar precipitation as covariate. The resulting field obtained by KED is filled of precipitation even over the sea where neither raingauges nor radar have measured precipitation. The result changes i

Re: [R-sig-Geo] Kriging prediction variances on gstat

Thanks! > Il giorno 21 giu 2017, alle ore 10:52, Edzer Pebesma > ha scritto: > > citation("gstat") > > points you to the literature. > >> On 21/06/17 10:16, Carlo Cavalieri wrote: >> Hi all, >> >> I would like to know how the kriging variance is computed by gstat, because >> on my analysis

Re: [R-sig-Geo] Kriging prediction variances on gstat

citation("gstat") points you to the literature. On 21/06/17 10:16, Carlo Cavalieri wrote: > Hi all, > > I would like to know how the kriging variance is computed by gstat, because > on my analysis I need to use a cell-specific covariate depended scale > parameter rather than a unique one. > If

[R-sig-Geo] Kriging prediction variances on gstat

Hi all, I would like to know how the kriging variance is computed by gstat, because on my analysis I need to use a cell-specific covariate depended scale parameter rather than a unique one. If the overall scale of the origins variance is fixed by the sill of the theoretical variogram model used

Re: [R-sig-Geo] Kriging in gstat - "number of rows in data.frame and SpatialPoints don't match" error

Hi Anna, The problem seems to be this line: > grd <- as(grid, "SpatialPixelsDataFrame") as the grid doesn't have any data except for the coordinates. Try instead with: > grd <- as(grid, "SpatialPixels") Cheers, Jon On 11/8/2016 3:50 PM, Anna Szyniszewska wrote: Hi, I have done kriging using

[R-sig-Geo] Kriging in gstat - "number of rows in data.frame and SpatialPoints don't match" error

Hi, I have done kriging using gstat a few times before, but this time I keep getting an error and I am unsure what is that the krige() function doesn't like - my spatial pixels or point data frames. Error in validObject(.Object) : invalid class "SpatialPointsDataFrame" object: number of rows i

Re: [R-sig-Geo] Kriging with uncertain data

Hi, Thank you for your suggestions. I gave the gstat with weights argument a try, but I’m not sure about the results / my implementation. I have worked out an example with the meuse dataset so I can share it with you: library(gstat) library(sp) data(meuse) coordinates(meuse) <- ~x+y data(meus

Re: [R-sig-Geo] Kriging with uncertain data

Dear Santiago, in this case I would interpolate/krige the uncertainty as well. Since uncertainty might have different distribution, different covariates, and different spatial autocorrelation than those of the measured value, I would build a new kriging model (fit new semivariogram, etc.) and

Re: [R-sig-Geo] Kriging with uncertain data

If the only problem is to krige these data, the solution is pretty trivial; add a location specific value to the nugget; this is what Delhomme in 1978 coined as regression kriging [1] (kriging of regressed rather than observed values, using estimates + estimation errors). An implementation is foun

Re: [R-sig-Geo] Kriging with uncertain data

Dear Ákos, I was referring to the former: I have data with two values at each location: measured value and uncertainty of the measurement. So, each observation is in fact a statistical variate, which we can assume is Gaussian distributed. Hence, my two values are the expected (mean) and the var

Re: [R-sig-Geo] Kriging with uncertain data

Dear Santiago, you mean you have two values at each location (observed value and uncertainty)? Or you have an observed value that is the sum of the real value and the observation error (uncertainty). If the last, then I think using the gstat::krige() function is straightforward, since the resu

[R-sig-Geo] Kriging with uncertain data

Dear R-sig-geo list members, I am curious about what are sensible approaches to spatial interpolation, most especially by using kriging, in the context of uncertain data. Suppose one has a dataset of values observed at different locations, and each value consists on the expected value and its v

[R-sig-Geo] Kriging - model selection

Dear all, I would like to perform interpolation of my data, using kriging. I am pretty new to this field, I would appreciate any advice. After reading various sources about how to do this kind of analysis in R, i came out with the code below. What I am struggling with, is the selection of the "co

[R-sig-Geo] Kriging with basis functions representation

(I am sorry if you have received this mail twice) Dear all,I have some spatial measurements over the area and I would like to make kriging to see the measurement points are more important points in the whole region. That would influence a lot the reported error on the predictions. I am trying to

[R-sig-Geo] kriging with non-euclidean distances

This is a recurring question from time to time. I finally packaged [1] some old code (2009) that modifies certain functions in geoR in order to use custom distance matrices in the construction of empirical variograms (variog), the fit of variogram models (likfit) and the kriging (krige.conv). I u

[R-sig-Geo] kriging to determine effect of area size on global Moran index

Dear bloggers, I am working with spatial polygons to see how variables are spatially correlated and how such correlation changes in time in a specific country. I want to look a bit more into detail at Moran’sI and how the morphology (surface and/or population) of the areas I consider can impact

[R-sig-Geo] Kriging: which and when

Dear List, I have a (maybe) strange question about interpolation methods. Just to make an example, let's suppose I have a SpatialPointsDataFrame of 100 meteorological station. Each point is a station with, for example, coordinates, mean annual temperature, elevation, distance from sea, slope and as

Re: [R-sig-Geo] Kriging

The kriging::kriging is using by default spherical-model. What kind of kriging function (or package) do you recommend? My data are reanalysis data form the ECMWF ( http://data-portal.ecmwf.int/data/d/interim_daily/levtype=sfc/) and the aim is to interpolate them. I would be grateful, if you co

Re: [R-sig-Geo] Kriging

You may have (near) duplicate points in your data. Other packages let you control more when you want to kriging, see the Spatial task view on CRAN. To me, kriging::kriging seems a bit of a black box: does it compute Euclidian distances from long/lat coordinates directly? How does it fit variogram

[R-sig-Geo] Kriging

Hello, I have a problem with the kriging function {kriging}. I get this error: "Error in solve.default(matrix(A, n + 1, n + 1)) : system is computationally singular: reciprocal condition number = 5.53457e-17" But it has worked with other data and now I don't understand why this error. Here is

Re: [R-sig-Geo] Kriging and datum shifts

Yes Edzer, I was not using terminology accurately. Apologies, R is new to me. Let me dig deeper if this simple example is needed. I will make sure if gstat kriging methods correctly use great circle distances correctly. Jeroen On 2013-05-28 12:15 PM, Edzer Pebesma wrote: Jeroen, sp does not ha

Re: [R-sig-Geo] Kriging and datum shifts

Jeroen, sp does not have anything to do with kriging, so your comment is confusing. sp does create spatial objects from data.frame's, and may do some sanity checks on coordinates in case they are longlat -- you have to specify they are longlat, though. If they are specified as longlat, and e.g. g

Re: [R-sig-Geo] Kriging and datum shifts

Thank you Mike, I have indeed all the info that I need in the SPDF: lat, lon, CRS. The irregular distribution of data is not projection-related, but is prohibitive to using simple rasterization operations because source data and target cells do not always overlap. Hence the need for kriging (o

Re: [R-sig-Geo] Kriging and datum shifts

For a simple approach I would just do on the raw coordinates in the SPDF. You could operate on the NetCDF file itself by reading from it directly as a matrix with vectors, but since you mention the SPDF why not just do this: x <- as.data.frame(spdf) ## figure out what "longitude" and "latitude" ar

[R-sig-Geo] Kriging and datum shifts

Dear list, I am converting NetCDF data, placed on a global grid with irregular latitude values, to a regular WGS84 grid using IDW. It works, but I have one remaining issue for which I can use advice. The original data is oriented -280, +80 longitude, and it needs to be interpolated to -180,

[R-sig-Geo] Kriging

Dear All, I was trying to understand the spatial variability of a variable over a region (variogram and kriging). I have attache the file for you to have a look if ONLY you have a time. I am not familiar how R is computing variogram and ultimately kriging. I read couple of papers and I had difficu

Re: [R-sig-Geo] Kriging problem

> Dear All, > >> >> >> I was trying to understand the spatial variability of a variable over a >> region (variogram and kriging). I have attache the file for you to have a >> look if ONLY you have a time. I am not familiar how R is computing >> variogram and ultimately kriging. I read couple of pa

[R-sig-Geo] Kriging problem

Dear All, I was trying to understand the spatial variability of a variable over a region (variogram and kriging). I have attache the file for you to have a look if ONLY you have a time. I am not familiar how R is computing variogram and ultimately kriging. I read couple of papers and I had difficu

[R-sig-Geo] kriging on transformed data

I am beginner in R.I have seen that prior to using kriging the data distribution has to be investigated to check if they are gaussian. The quantile-quantile plot of rainfall on a day using 50 stations data . The q-q plotof log

[R-sig-Geo] Kriging of values distributed on line

Hello, perhaps I am asking for something obvious, but I would like to perform interpolation of values, which are distributed on the line. my dataset consists of two vectors: stationing (distance from origin) and values. I would like to krige the values to obtain smooothed curve with variance estim

Re: [R-sig-Geo] kriging limits

with country boundaries but I think you can easily melt these > using rgeos:::gUnaryUnion > > Thanks, > > Simon > > -Original Message- > From: r-sig-geo-boun...@r-project.org [mailto: > r-sig-geo-boun...@r-project.org] On Behalf Of marco milella > Sent: 29

Re: [R-sig-Geo] kriging limits

al Message- From: r-sig-geo-boun...@r-project.org [mailto:r-sig-geo-boun...@r-project.org] On Behalf Of marco milella Sent: 29 October 2012 22:12 To: r-sig-geo Subject: Re: [R-sig-Geo] kriging limits Dear Rob, thanks for the help. However, my shp file is a SpatialLinesDataFrame, so things not

Re: [R-sig-Geo] kriging limits

Dear Rob, thanks for the help. However, my shp file is a SpatialLinesDataFrame, so things not work (blue plot). On the other hand, I tried with an old map of Europe (SpatialPolygonsDataFrame) and it works fine. Do any of you know where to find SpatialPolygonsDataframe .shp file of all world (possi

[R-sig-Geo] Kriging on Ecological data

Hi All, I had some species count of a particular area. with the corresponding plot numbers and location info. Anyone have idea on how to do Kiriging(any type) on it Sample data format: Latitute, Longitude, PlotID, SpeciesCount 18.5 32.3 105 100 ... .. Please tell me h

Re: [R-sig-Geo] kriging limits

Marco, 'colouring' only the land is relatively easy, kriging over just the land is much harder... To do the former, all you need to to do is plot your kriged map, then plot a 'mask' over the sea area. To create the mask, you need something like the snippet below world<-readShapeSpatial("10m_coast

[R-sig-Geo] kriging limits

Dear all, I'm a relatively new with R. However, I realized from colleagues the great capabilities of R or analyzing geo data. I'm interested in performing an ordinary kriging to interpolate z value on a world map sarting from a dataset that (unfortunately) includes rather few observations (13 loca

[R-sig-Geo] kriging log transformed data

Hi, I am fairly new to R and gstat and kriging. However, I have been trying to use gstat to carry out kriging on a set of heavy metal data that I have. The heavy metal values are not normally distributed, however, a log tranformation can get the data to be normal. There is also some correlation be

Re: [R-sig-Geo] Kriging with known, varying measurement errors

You should be able to do this with the weights argument of gstat, have a look at this posting that answers a similar question: http://geostatistics.forum.52north.org/Re-question-about-gstat-td2889218.html Jon On 12-Jul-12 18:02, cmk24 wrote: I am trying to use R to krige data that has known, v

[R-sig-Geo] Kriging with known, varying measurement errors

I am trying to use R to krige data that has known, varying measurement errors. From the gstat user manual it seems like this should be possible, I just can't figure out hot to implement it. Does anyone know how I can do this? -- View this message in context: http://r-sig-geo.2731867.n2.nabble.c

Re: [R-sig-Geo] kriging with external drift problems

Dear Els, I believe there is no cure for too little data. Instead of looking at directional variograms, I would look at variograms of regression residuals. Instead of a max distance value, I would set a maximum neighbourhood size expressed as the max number of nearest values. Whether 43 observati

[R-sig-Geo] kriging with external drift problems

Dear list, I have a dataset from Ethiopia with a good correlation between infiltration depth and slope (r=0.75). Therefore, I would like to use KED (with slope as secondary variable) to interpolate the infiltration observations. However, my main problem is the limited amount of observations (n

[R-sig-Geo] kriging and second order trend

I would like to remove a second order trend from elevation data before kriging. Was thinking on using surf.ls to fit the trend but not sure how to pass it onto the krige argument. Any suggestions? Thanks [[alternative HTML version deleted]] __

Re: [R-sig-Geo] Kriging wiht anisotropy modeling using great circle distance in gstat [SEC=UNCLASSIFIED]

regards, Jin -Original Message- From: r-sig-geo-boun...@r-project.org [mailto:r-sig-geo-boun...@r-project.org] On Behalf Of Johan Van de Wauw Sent: Saturday, 20 August 2011 12:49 AM To: r-sig-geo@r-project.org Subject: Re: [R-sig-Geo] Kriging wiht anisotropy modeling using great circle

Re: [R-sig-Geo] Kriging wiht anisotropy modeling using great circle distance in gstat [SEC=UNCLASSIFIED]

On Fri, Aug 19, 2011 at 3:37 AM, wrote: > Error in predict.gstat(g, newdata = newdata, block = block, nsim = nsim,  : >  gstat: value not allowed for: for long/lat data, anisotropy cannot be defined >> > > Is there anything wrong in my code or it is a bug? Any suggestions are > appreciated! >

[R-sig-Geo] Kriging wiht anisotropy modeling using great circle distance in gstat [SEC=UNCLASSIFIED]

Hi All, I tried to do kriging with anisotropy using great circle distance as bellow. > library(gstat) > loadMeuse() > > require(rgdal) > proj4string(meuse) = CRS("+init=epsg:28992") > meuse.ll = spTransform(meuse, CRS("+proj=longlat")) > > data(meuse.grid) > coordinates(meuse.grid) = ~x+

[R-sig-Geo] kriging variance vs sample spacing

Hi I want to compute the kriging variance as a function of grid spacing so that I can measure cost vs. kriging variance in deciding on an appropriate grid spacing for an ocean sediment surrounding an outfall. From an earlier study I have a response with an anisotropic variogram given by model=

Re: [R-sig-Geo] kriging and cokriging with multiple covariate

Dear Andrew, Thank you very much Toni - Original Message - From: "Andrew Finley" To: "TM" Cc: Sent: Tuesday, February 08, 2011 1:09 PM Subject: Re: [R-sig-Geo] kriging and cokriging with multiple covariate Hi Toni, Check out the spMvLM function in the

Re: [R-sig-Geo] kriging and cokriging with multiple covariate

Hi Toni, Check out the spMvLM function in the spBayes package (or spMvGLM depending on your outcome variable), there are also some illustrative slides and examples toward the bottom of this page http://blue.for.msu.edu/JBC_10/SC . Let me know if you have any questions- Andy Quoting TM :

[R-sig-Geo] kriging and cokriging with multiple covariate

Hi all, I'm trying to model some ecological data of an area to predict spatially some ecological response variable in terms of x, y, and some covariates have been measured. The problem is that some of the covariates are discrete (ordinal) and do not really know how to make a cokriging with mul

Re: [R-sig-Geo] kriging function

On 12/12/2010 08:48 AM, Nikki roy wrote: > " > any suggestions? > Yes, provide us with an example that we can reproduce, ideally based on the meuse + meuse.grid data set, where you could introduce some missing values. -- Edzer Pebesma Institute for Geoinformatics (ifgi), University of Münster W

[R-sig-Geo] kriging function

Dear All, I am doing regression kriging of a small dataset."How can i stop krige() function from deleting the NA columns in my input data" 1. # define the linear model by ordinary least squares prediction pb.ols <- lm(pb ~ geul$eucdist + geul$elev, data=geul) 2. # add residual to geul geul$res =