Package ‘SSN’ February9,2023 Type Package Title SpatialModelingonStreamNetworks Version 1.1.16 Date 2023-02-09 Depends R(>=4.0.0),RSQLite(>=1.1-2),sp Imports MASS,igraph(>=1.0.0),maptools,lattice,methods,Matrix, rgdal(>=1.2-5),rgeos(>=0.3-22) Maintainer JayVerHoef<[email protected]> Description Spatialstatisticalmodelingandpredictionfordataonstreamnetworks,includingmod- elsbasedonin-streamdistance(VerHoef,J.M.andPeter- son,E.E.,2010.<DOI:10.1198/jasa.2009.ap08248>.)Modelsarecreatedusingmovingaver- ageconstructions.Spatiallinearmodels,includingexplanatoryvariables,canbefitwith(re- stricted)maximumlikelihood. Mappingandothergraphicalfunctionsareincluded. License GPL-2 LazyLoad yes LinkingTo BH NeedsCompilation yes Author JayVerHoef[aut,cre], ErinPeterson[aut] Repository CRAN Date/Publication 2023-02-0920:00:06UTC R topics documented: SSN-package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 additive.function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 AIC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 as.SpatialLines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 binSp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 BlockPredict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 BLUP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1 2 Rtopicsdocumented: boxplot.SpatialStreamNetwork . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 copyLSN2temp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 covparms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 createDistMat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 createSSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 CrossValidationSSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 CrossValidationStatsSSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Designfunctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 EmpiricalSemivariogram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 fitNS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 fitRE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 fitSimBin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 fitSimGau . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 fitSimPoi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 fitSp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 fitSpBk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 fitSpRE1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 fitSpRE2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 getConfusionMat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 getPreds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 getSSNdata.frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 getStreamDistMat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 glmssn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 glmssn-class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 GR2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 importPredpts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 importSSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 influenceSSN-class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 InfoCritCompare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 mf04 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 mf04p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 MiddleFork04.ssn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 plot.glmssn.predict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 plot.influenceSSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 plot.SpatialStreamNetwork . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 plot.Torgegram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 poiSp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 predict.glmssn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 print.summary.glmssn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 putSSNdata.frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 residuals.glmssn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 SimulateOnSSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 SpatialStreamNetwork-class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 splitPredictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 subsetSSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 summary.glmssn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Torgegram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 updatePath. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 SSN-package 3 varcomp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 writeSSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Index 95 SSN-package SpatialModelingonStreamNetworks Description CreatesspatialstreamnetworkrepresentationsinRandfitsspatialmodels. Details Package: SSN Type: Package Version: 1.1.12 Date: 2018-01-24 License: GPL-2 LazyLoad: yes The SSN package provides tools to fit generalized linear models with spatial autocorrelation for streamnetworkdata. SSNusesnormallikelihoodmethods(includingREML)andquasi-likelihood forPoissonandBinomialfamilies. ThespatialformulationwasoriginallydescribedinVerHoef, Peterson, and Theobald (2006), with more details given by Ver Hoef and Peterson (2010) and Peterson and Ver Hoef (2010). The spatial data must be formatted in a geographic information system(GIS)priortoimportingitintoR.AcustomArcGIStoolboxhasbeenprovidedtoformat thedata: SpatialToolsfortheAnalysisofRiverSystems(STARS)toolset(PetersonandVerHoef, 2014),andtheSSNpackageisfullydescribedinVerHoef,Peterson,Clifford,andShah(2014). Author(s) JayVerHoefandErinPeterson<[email protected]> References VerHoef,J.M.,Peterson,E.E.andTheobald,D.(2006)SpatialStatisticalModelsthatUseFlow andStreamDistance. EnvironmentalandEcologicalStatistics13,449–464. Ver Hoef, J. M. and Peterson, E. E. (2010) A Moving Average Approach for Spatial Statistical Models of Stream Networks (with Discussion). Journal of the American Statistical Association 105,6–18. DOI:10.1198/jasa.2009.ap08248. Rejoinderpgs. 22–24. Peterson,E.E.andVerHoef,J.M.(2010)AMixed-ModelMoving-AverageApproachtoGeosta- tisticalModelinginStreamNetworks. Ecology91(3),644–651. Peterson,E.E.andVerHoef,J.M.2014.STARS:AnArcGISToolsetUsedtoCalculatetheSpatial InformationNeededtoFitSpatialStatisticalModelstoStreamNetworkData. JournalofStatistical Software56(2): 1–17. 4 additive.function VerHoef,J.M.,Peterson,E.E.,Clifford,D.andShah,R.(2014)SSN:AnRPackageforSpatial StatisticalModelingonStreamNetworks. JournalofStatisticalSoftware56(3): 1–45. additive.function GenerateanAdditiveFunctionValue Description Generate an additive function value based on a proportional influence variable into an additive functionvalue Usage additive.function(ssn, VarName, afvName) Arguments ssn aSpatialStreamNetwork-classobject VarName Thenameofthethevariablethatwillbeusedtocalculatetheadditivefunction value. Thedata.framessn@datamustcontainacolumnwiththisname. afvName Thenameassignedtothecolumnofadditivefunctionvalues,whichareadded to the ssn@data data.frame object, as well as the data.frames for the observed andpredictionsites. Details Calculatingtheadditivefunctionvalues(AFVs)isatwostepprocess;firsttheVarNamevaluesare usedtocalculatethesegmentproportionalinfluences(PIs).ThenthesegmentPI’sareusedtocalcu- latetheAFVsforeachlinesegment,observedsite,andpredictionsiteintheSpatialStreamNetwork- class object. A detailed description of the segment PIs and the steps used to calculate AFVs are provided in Peterson and Ver Hoef (2010; Appendix A). The AFVs can also be calculated using theSpatialToolsfortheAnalysisofRiverSystems(STARS),whichisacustomArcGIS(version 9.3.1)toolbox. Value TheSpatialStreamNetworkobject,ssn,withanewcolumnnamedVarNameincludedinthedata.frames forthelines,observedsites,andpredictionsitestoholdtheAFVs. Author(s) RohanShah<[email protected]> References Peterson,E.E.andVerHoef,J.M.(2010)Amixed-modelmoving-averageapproachtogeostatis- ticalmodelinginstreamnetworks. Ecology91(3),644–651. Peterson E.E.(2011)STARS: Spatial Tools for the Analysis of River Systems: A tutorial. CSIRO TechnicalReportEP111313. 42p. AIC 5 Examples library(SSN) #for examples, copy MiddleFork04.ssn directory to R's temporary directory copyLSN2temp() # NOT RUN # Create a SpatialStreamNetork object that also contains prediction sites #mf04p <- importSSN(paste0(tempdir(),'/MiddleFork04.ssn'), # predpts = "pred1km", o.write = TRUE) #use mf04p SpatialStreamNetwork object, already created data(mf04p) #for examples only, make sure mf04p has the correct path #if you use importSSN(), path will be correct mf04p <- updatePath(mf04p, paste0(tempdir(),'/MiddleFork04.ssn')) #Calculate an additive function value based on an existing column. names(mf04p@data) mf04p <- additive.function(mf04p, "h2oAreaKm2", "areaAFV") #notice that a column called afvArea was already included, and "areaAFV" replicates it # in the lines data head(mf04p@data) # and in the observed points data head(getSSNdata.frame(mf04p)) # and in the prediction points data head(getSSNdata.frame(mf04p, "pred1km")) AIC AICforglmssnobjects Description AIC.glmssnisamethodthatcalculatesAICforfittedglmssnobjects. Usage ## S3 method for class 'glmssn' AIC(object, ..., k = 2) Arguments object anobjectofclassglmssn ... optionallymorefittedmodelobjects k numeric,thepenaltyperparametertobeused;thedefaultk=2istheclassical AIC. Details AICisagenericfunctionandthisimplementsamethodforglmssnobjects 6 as.SpatialLines Value anumericAICvalueforthespecifiedglmssnobject Author(s) JayVerHoef<[email protected]> SeeAlso glmssn Examples library(SSN) #for examples, copy MiddleFork04.ssn directory to R's temporary directory copyLSN2temp() # NOT RUN # Create a SpatialStreamNetork object that also contains prediction sites #mf04p <- importSSN(paste0(tempdir(),'/MiddleFork04.ssn'), # predpts = "pred1km", o.write = TRUE) #use mf04p SpatialStreamNetwork object, already created data(mf04p) #for examples only, make sure mf04p has the correct path #if you use importSSN(), path will be correct mf04p <- updatePath(mf04p, paste0(tempdir(),'/MiddleFork04.ssn')) # get some model fits stored as data objects data(modelFits) #NOT RUN use this one #fitSp <- glmssn(Summer_mn ~ ELEV_DEM + netID, # ssn.object = mf04p, EstMeth = "REML", family = "Gaussian", # CorModels = c("Exponential.tailup","Exponential.taildown", # "Exponential.Euclid"), addfunccol = "afvArea") #for examples only, make sure fitSp has the correct path #if you use importSSN(), path will be correct fitSp$ssn.object <- updatePath(fitSp$ssn.object, paste0(tempdir(),'/MiddleFork04.ssn')) #note the model was fitted using REML, so fixed effects have # been integrated out summary(fitSp) AIC(fitSp) as.SpatialLines MethodstoconvertSpatialStreamNetworkobjectsclassestospclasses as.SpatialLines 7 Description ConvertsSpatialStreamNetworkobjectstospobjects. Usage ## S3 method for class 'SpatialStreamNetwork' as.SpatialLines(x, ...) ## S3 method for class 'SpatialStreamNetwork' as.SpatialPoints(x, data = "Obs", ...) ## S3 method for class 'SpatialStreamNetwork' as.SpatialLinesDataFrame(x, ...) ## S3 method for class 'SpatialStreamNetwork' as.SpatialPointsDataFrame(x, data = "Obs", ...) Arguments x an SpatialStreamNetwork object to be converted to class SpatialLines, Spa- tialPoints,SpatialLinesDataFrameorSpatialPointsDataFramefromthesppack- age. data thedatasetintheSpatialStreamNetworkobjecttoconvert.TheSpatialStreamNetwork objectcanholdmultiplespatialpointdatasets,includingtheobserveddataand multiplepredictiondatasets. SeeSpatialStreamNetwork-class. ... optionalargumentsforspecificmethodswrittenforthesegenerics Value as.SpatialLines.SpatialStreamNetwork converts an object of class SpatialStreamNetwork toanobjectofclassSpatialLinesfromthesppackage,as.SpatialPoints.SpatialStreamNetwork converts an object of class SpatialStreamNetwork to an object of class SpatialPoints from thesppackage,andas.SpatialPointsDataFrame.SpatialStreamNetworkconvertsanobjectof classSpatialStreamNetworktoanobjectofclassSpatialPointsDataFramefromthesppack- age, Author(s) JayVerHoef<[email protected]> SeeAlso spplot Examples library(SSN) #for examples, copy MiddleFork04.ssn directory to R's temporary directory copyLSN2temp() # NOT RUN # Create a SpatialStreamNetork object that also contains prediction sites 8 binSp #mf04p <- importSSN(paste0(tempdir(),'/MiddleFork04.ssn'), # predpts = "pred1km", o.write = TRUE) #use mf04p SpatialStreamNetwork object, already created data(mf04p) #for examples only, make sure mf04p has the correct path #if you use importSSN(), path will be correct mf04p <- updatePath(mf04p, paste0(tempdir(),'/MiddleFork04.ssn')) names(mf04p) #--------- # make plots using sp methods #--------- #plot the stream lines plot(as.SpatialLines(mf04p), col = "blue") # add the observed locations with size proportional # to mean summer temperature plot(as.SpatialPoints(mf04p), pch = 19, cex = as.SpatialPointsDataFrame(mf04p)$Summer_mn/9 , add = TRUE) # add the prediction locations on the 1 km spacing plot(as.SpatialPoints(mf04p, data = "pred1km"), cex = 1.5, add = TRUE) # add the dense set of points for block prediction on Knapp segment plot(as.SpatialPoints(mf04p, data = "Knapp"), pch = 19, cex = 0.3, col = "red", add = TRUE) binSp FittedglmssnobjectforexampledatasetMiddleFork.ssn Description TheMiddleFork04.ssndatafoldercontainsthespatial,attribute,andtopologicalinformationneeded toconstructaspatialstreamnetworkobjectusingtheSSNpackage. Thisisafittedmodelusingthe glmssnfunction. Details Seethehelpforglmssnforhowthemodelwascreated. Examples library(SSN) data(modelFits) ls() BlockPredict 9 BlockPredict BlockPredictonforStreamsData Description Blockpredictionforobjectsofclassglmssn-class Usage BlockPredict(object, predpointsID) Arguments object anobjectofclassglmssn predpointsID avalidpredictionpointsID Details Thisfunctionoperatesonglmssnobjectsinmuchthesamewayasthepredictfunction.BlockPredict usesthelocationsinthepredpointsIDdatasettocomputetheaveragepredictionvalueinthearea definedbythepredictionlocations. Thesepredictionlocationsareusedtoapproximatetheintegral overthatarea,sotheyshouldbeevenlyspacedanddenseintheareawhereblockpredictionisde- sired. TheuserneedstocreatethesepredictionlocationsandincludethemintheSSNobjectprior tofittingthemodelwithglmssn. Value Adata.framewithonerowandtwocolumns. Thefirstcolumn, BlockPredEst, istheaveragepre- dictionvalue,andthesecondcolumn,BlockPredSE,isthestandarderroroftheblockprediction. Author(s) JayVerHoef<[email protected]> References Ver Hoef, J. M.. Peterson, E. E. and Theobald, D. (2006) Spatial statistical models that use flow andstreamdistance. EnvironmentalandEcologicalStatistics13,449-464. DOI:10.1007/s10651- 006-0022-8. Examples ## Not run: library(SSN) #for examples, copy MiddleFork04.ssn directory to R's temporary directory copyLSN2temp() # NOT RUN 10 BLUP # Create a SpatialStreamNetork object that also contains prediction sites #mf04p <- importSSN(paste0(tempdir(),'/MiddleFork04.ssn'), # predpts = "pred1km", o.write = TRUE) #use mf04p SpatialStreamNetwork object, already created data(mf04p) #for examples only, make sure mf04p has the correct path #if you use importSSN(), path will be correct mf04p <- updatePath(mf04p, paste0(tempdir(),'/MiddleFork04.ssn')) # Not needed: already added, # add densely gridded prediction points for two stream segments # mf04p <- importPredpts(mf04p, "Knapp", "ssn") # mf04p <- importPredpts(mf04p, "CapeHorn", "ssn") names(mf04p) # see densely gridded prediction points on stream plot(mf04p, PredPointsID = "Knapp") # you could fit the model #fitSpBk <- glmssn(Summer_mn ~ ELEV_DEM + netID, # ssn.object = mf04p, EstMeth = "REML", family = "Gaussian", # CorModels = c("Exponential.tailup","Exponential.taildown", # "Exponential.Euclid"), addfunccol = "afvArea") # or load the pre-fit model data(modelFits) fitSpBk$ssn.object <- updatePath(fitSpBk$ssn.object, paste0(tempdir(),'/MiddleFork04.ssn')) # one-at-a-time predictions for CapeHorn stream ## NOTE: need the amongpreds distance matrices for block prediction createDistMat(mf04p, predpts = "CapeHorn", o.write = TRUE, amongpreds = TRUE) fitSpPredC <- predict(fitSpBk, "CapeHorn") # plot densely gridded prediction points on stream plot(fitSpPredC, "Summer_mn") # block prediction for CapeHorn stream BlockPredict(fitSpBk, "CapeHorn") ## Another example # one-at-a-time predictions for Knapp stream createDistMat(mf04p, predpts = "Knapp", o.write = TRUE, amongpreds = TRUE) fitSpPredK <- predict(fitSpBk, "Knapp") # plot densely gridded prediction points on stream plot(fitSpPredK, "Summer_mn") # block prediction for Knapp stream BlockPredict(fitSpBk, "Knapp") ## End(Not run)