Unveiling the influence of climatic factors on reproductive patterns in neotropical bats through a Bayesian framework

Author

TESSIER/GRAVE/DAVID/DENOUAL/SERRE/LAPORTE/LEROUX/BRUZZO

Published

October 22, 2024

I. Some history about bat studies in French Guiana

French Guiana is a territory belonging to the bio-geographic unit of the Guiana Shield (Hollowell and Reynolds (2005)) With an area of 83,846 km², it hosts, like its neighboring territories, remarkable species diversity. Covered by 97% forest habitat mosaic, its extreme species diversity is well-documented. French Guiana represents a global hotspot in terms of bat species richness. The most commonly used method for studying bats in tropical regions is ground-level mist netting, followed by canopy netting and direct roost searche (MacSwiney G., Clarke, and Racey (2008)). It is with these methods that the first studies on bat community characterization were conducted (Charles-Dominique Pierre, Brosset André, and Jouard Sylvie (2001)). To date, very little work has focused on exploring the reproduction patterns of Neotropical bat communities. Reproductive patterns are closely related to rainfall regimes (Ruiz-Ramoni, Ramoni-Perazzi, and Munoz-Romo (2016)) frequently occurring twice a year in tropical regions.

The data collected on the species of bats captured by date and location in French Guiana, and containing information on sex, age and reproductive state
The meteorological data of French Guiana from 1950 to 2024
The capture locations coordinates (degree minute second)

II. Formating our datasets

A. Importation of datasets and packages

library(dplyr)
library(readxl)
library(lubridate)
library(ggplot2)
library(stringr)
library(tidyr)
library(R2jags)
library(ade4)
library(spdep)
library(tripack)
library(sp)
library(geoR)

data_chiro<-read_excel("data_chiro.xlsx")
data_abiot_50_22<-read.csv("abiot_1950_2022.csv", sep = ";") 
data_abiot_23_24<-read.csv("abiot_2023_2024.csv", sep = ";") 
coord_chiro<-read_excel("COORD_CAPTURE.xlsx")

B. Processing of the abiotic dataset

Firstly we concatenate all the abiotic data together
Then we start to clean the abiotic the dataset by deleting all years that precede the first event of capture, the we use the Lubridate R package (Spinu, Grolemund, and Wickham (2010)) to formate the data column as we want and the dplyr R package (Wickham et al. (2023)) to reorganize the data set.

data_abiot<-rbind(data_abiot_50_22, data_abiot_23_24)
data_abiot <- data_abiot %>%
  filter(AAAAMMJJ >= 20141028) %>% 
  mutate(AAAAMMJJ = ymd(AAAAMMJJ)) %>% 
  mutate(Year = year(AAAAMMJJ)) %>% 
  mutate(AAAAMMJJ = format(AAAAMMJJ, "%d-%m-%Y")) %>% 
  select(NUM_POSTE, NOM_USUEL, LAT, LON, AAAAMMJJ, RR, TM, TAMPLI, FFM, Year) %>%
  rename(date = AAAAMMJJ, rain_mm = RR, Temperature = TM, ampli_Temp = TAMPLI, wind_ms = FFM, COMMUNE = NOM_USUEL)%>% 
  mutate(julian_day = yday(date))

Computing moonlight intensity :

Function that simulates the lunar cycle as a proxy for night-time luminosity

Lune <- function(date){
  sortie <- c() 
  for(i in 1:length(date)){ 
    date_num <- as.numeric(as.Date(date[i],format="%d-%m-%Y")) 
    sortie[i] <- (0.5*(cos((date_num-21)*((2*pi)/(29.530589)))+1)) 
  }
  return(sortie)
}

data_abiot$Lune <-Lune(data_abiot$date)

C. Processing of the biotic dataset

Data_chiro table pre-processing : We only keep the columns of interest, then we only keep the female individuals because the reproduction indices are mainly observable on females. We only keep adult individuals because this is the age class likely to have reproductive indices.

data_chiro <- data_chiro %>%
  select(c(Année, Commune, Localité, Date, Espèce, Sexe, Repro, Age)) %>% #select the columns we want
  filter(Sexe == "F") %>% 
  filter(Age == "AD") %>% 
  filter(is.na(Repro)*1==0)%>%
  mutate(Date = ymd(Date)) %>%  
  mutate(Year = year(Date)) %>% 
  mutate(Date = format(Date, "%d-%m-%Y"))%>% 
  mutate(julian_day = yday(Date))%>% #Date =julian_day
  rename(LIEU_DIT = Localité, COMMUNE = Commune) #on renomme

Converting DMS coordinates to decimal coordinates :

Function to converting minute/second degree into decimal degrees

convert_dms_to_decimal <- function(dms) {
  parts <- str_match(dms, "(\\d+)°(\\d+)'(\\d+\\.?\\d*)''\\s([NSEW])")
  degrees <- as.numeric(parts[, 2]) 
  minutes <- as.numeric(parts[, 3])
  seconds <- as.numeric(parts[, 4])
  direction <- parts[, 5]
  decimal <- degrees + minutes / 60 + seconds / 3600
  if (direction %in% c("S", "W")) {
    decimal <- -decimal
  }
  
  return(decimal)
}

We convert the DMS coordinates to decimal coordinates and we merge it with data_chiro by the capture locality

library(stringr)
coord_chiro <- coord_chiro %>%
  rowwise() %>%  
  mutate(
    lon_dms = str_extract(COORD, "^[^N]+W"), 
    lat_dms = str_extract(COORD, "[^W]+N"),  
    Longitude = convert_dms_to_decimal(lon_dms), 
    Latitude = convert_dms_to_decimal(lat_dms) 
  ) %>%
  ungroup() %>% 
  select(COMMUNE, LIEU_DIT, Latitude, Longitude)%>%
  rename(LAT = Latitude, LON = Longitude)

data_chiro <- data_chiro %>%
  inner_join(coord_chiro, by = "LIEU_DIT") 

data_chiro <- data_chiro%>%
  select(COMMUNE.x,LIEU_DIT,Date,Espèce,Sexe,Repro,Age,Year,julian_day,LAT,LON)%>% 
  rename(COMMUNE = COMMUNE.x)

D. Graphical representation of French Guiana with the incorporation of meteorological stations and capture points

In the graphic we can see in blue the capture points and in green the meteorological stations

limite <- matrix(c(-54.40582467596958,5.123666986130482,0 -54.45958864979806,4.753175826159146,0 -54.43393543693306,4.077331065323269,0 -54.01198973780399,3.581399050490593,0 -54.20921816638264,3.132878544854336,0 -54.21013685965725,2.796587485871633,0 -54.45803260870573,2.441977197957256,0 -54.68709817836254,2.314479358976484,0 -54.15543583603689,2.123868408305178,0 -53.7651653142448,2.30335873628299,0 -53.35415060581271,2.164099707075567,0 -52.94875879245732,2.132858726757194,0 -52.56611638847146,2.518366090728264,0 -52.59199025844659,2.634734023219086,0 -52.39229001432994,2.890167168832993,0 -52.37091318927082,3.144918005801542,0 -52.21064002801557,3.274351081759725,0 -51.9836274889358,3.696015530721377,0 -51.6721839730206,4.031282773145124,0 -51.63663780527241,4.28966735155517,0 -51.9023261931293,4.501855027711033,0 -52.05244725433236,4.818896971774704,0 -52.51627884489513,5.026927311475695,0 -52.703088477061,5.197262336288941,0 -53.15061905132664,5.577522799535109,0 -53.53420039840946,5.573574890423137,0 -53.95403768383684,5.781472322818466,0 -54.40582467596958,5.123666986130482),ncol=2,byrow=T)

plot(limite[,1],limite[,2],type='l')
points(data_abiot$LON,data_abiot$LAT,pch=20,col='green')
points(coord_chiro$LON,coord_chiro$LAT,pch=20,col="blue")

To estimate the meteorogical values at capture sites, we used spatial interpolation. For each date and each value of interest, we measured at first wether or not there was an spatial structure using the moran test. If there is no spatial structure, we estimate the value by meaning the observed values. If there is a spatial structure, we realise a kriging. Because of the necessity to automatise this interpolation, the method is not completely correct, however the weather in French Guiana is not very heterogenous, and is mostly gradients to the ocean or east-west gradient. Morover, the data we obtain show the same patterns of dry season and rain season, and for all the values.

datamap<-read.csv("data_meteo.csv",row.names = 1)
coord <- datamap[,c(4,3)]

dd <- unique(data_chiro$Date)
res<- c()
for(d in dd){
  
  if(length(datamap$date[which(datamap$date==d)])>0){ 
    coord_d <- coord[which(datamap$date==d),]
    delau <- c()
    delau<-tryCatch({ tri2nb(coord_d)}, error=function(e){}) 
    delau.w<- nb2listw(delau, style="S", zero.policy=TRUE)
    m<-c()
    m <- moran.test(datamap$rain_mm[which(datamap$date==d)], delau.w, zero.policy=TRUE) 
    if(m$p.value < 0.05){
      spatial_point <- c()
      spatial_point <- SpatialPointsDataFrame(coord_d, as.data.frame(datamap$rain_mm[which(datamap$date==d)]))
      geodata <- as.geodata(spatial_point)
      coord <- datamap[,c(4,3)]
      variograme <- variog(geodata, option  = "sm")
      m.d <- max(variograme$u)
      if(length(which(is.na(variograme$v)==T))>0){m.d=variograme$u[which(is.na(variograme$v))][1]}
      variograme <- variog(geodata, option  = "sm",max.dist = m.d)
      mod1 <- variofit(variograme,ini.cov.pars = c((0.9*max(variograme$v,na.rm = T)[1]-variograme$u[1]),variograme$u[which(variograme$v==max(variograme$v,na.rm = T)[1])][1]),nugget=variograme$v[1],"exponential")#on réalise une interpolation spatiale 
      res[which(data_chiro$Date==d)] <- as.numeric(krige.conv(geodata, loc = data_chiro[which(data_chiro$Date==d),c(11,10)] , krige = krige.control(obj.m = mod1))$predict)
    }else{
      res[which(data_chiro$Date==d)] <- mean(datamap$rain_mm[which(datamap$date==d)])
    }   
    
  }else{
    res[which(data_chiro$Date==d)] <- NA
  }
  print(d)
  
}

[1] "05-02-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-03-2016"
[1] "06-07-2016"
[1] "13-09-2016"
[1] "15-09-2016"
[1] "25-10-2016"
[1] "26-10-2016"
[1] "04-01-2017"
[1] "25-03-2017"
[1] "26-03-2017"
[1] "27-03-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-04-2017"
[1] "17-04-2017"
[1] "28-05-2017"
[1] "17-06-2017"
[1] "20-06-2017"
[1] "06-07-2017"
[1] "23-07-2017"
[1] "01-08-2017"
[1] "10-08-2017"
[1] "12-08-2017"
[1] "16-08-2017"
[1] "04-09-2017"
[1] "14-09-2017"
[1] "27-09-2017"
[1] "27-10-2017"
[1] "28-10-2017"
[1] "30-10-2017"
[1] "31-10-2017"
[1] "01-11-2017"
[1] "25-11-2017"
[1] "09-12-2017"
[1] "19-01-2018"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-02-2018"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-02-2018"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-02-2018"
[1] "17-02-2018"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-05-2019"
[1] "16-05-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-05-2019"
[1] "20-05-2019"
[1] "28-05-2019"
[1] "31-05-2019"
[1] "05-06-2019"
[1] "06-06-2019"
[1] "12-06-2019"
[1] "14-06-2019"
[1] "19-06-2019"
[1] "21-06-2019"
[1] "27-06-2019"
[1] "03-07-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-07-2019"
[1] "10-07-2019"
[1] "12-07-2019"
[1] "17-07-2019"
[1] "30-07-2019"
[1] "01-08-2019"
[1] "05-08-2019"
[1] "14-08-2019"
[1] "18-08-2019"
[1] "22-08-2019"
[1] "02-09-2019"
[1] "05-09-2019"
[1] "12-09-2019"
[1] "18-09-2019"
[1] "21-09-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-09-2019"
[1] "07-10-2019"
[1] "08-10-2019"
[1] "16-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-10-2019"
[1] "26-10-2019"
[1] "27-10-2019"
[1] "28-10-2019"
[1] "01-11-2019"
[1] "03-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-11-2019"
[1] "05-11-2019"
[1] "06-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-11-2019"
[1] "18-11-2019"
[1] "20-11-2019"
[1] "22-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-06-2020"
[1] "11-10-2020"
[1] "07-11-2020"
[1] "08-11-2020"
[1] "09-11-2020"
[1] "10-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-11-2020"
[1] "12-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-01-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-04-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-07-2021"
[1] "02-10-2021"
[1] "16-10-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-11-2021"
[1] "28-11-2021"
[1] "06-12-2021"
[1] "07-12-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-12-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-02-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-03-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-03-2022"
[1] "03-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-04-2022"
[1] "11-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-04-2022"
[1] "25-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-04-2022"
[1] "26-05-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-06-2022"
[1] "22-06-2022"
[1] "30-06-2022"
[1] "08-07-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "09-07-2022"
[1] "20-07-2022"
[1] "06-08-2022"
[1] "19-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-08-2022"
[1] "27-08-2022"
[1] "28-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "29-08-2022"
[1] "02-09-2022"
[1] "09-09-2022"
[1] "13-09-2022"
[1] "19-09-2022"
[1] "01-10-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-11-2022"
[1] "04-11-2022"
[1] "05-11-2022"
[1] "15-11-2022"
[1] "28-01-2023"
[1] "29-01-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-02-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-03-2023"
[1] "17-03-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-03-2023"
[1] "20-04-2023"
[1] "23-04-2023"
[1] "27-04-2023"
[1] "06-05-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-05-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "29-05-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-05-2023"
[1] "31-05-2023"
[1] "01-06-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-06-2023"
[1] "26-06-2023"
[1] "27-06-2023"
[1] "28-06-2023"
[1] "08-09-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-09-2023"
[1] "27-09-2023"
[1] "28-09-2023"
[1] "30-09-2023"
[1] "01-10-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-10-2023"
[1] "07-10-2023"
[1] "13-10-2023"
[1] "14-10-2023"
[1] "20-10-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-10-2023"
[1] "27-10-2023"
[1] "28-10-2023"
[1] "30-10-2023"
[1] "31-10-2023"
[1] "01-11-2023"
[1] "02-11-2023"
[1] "09-11-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-11-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-12-2023"
[1] "20-12-2023"
[1] "10-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-02-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-02-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-02-2024"
[1] "21-02-2024"
[1] "05-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-03-2024"
[1] "28-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-03-2024"
[1] "06-04-2024"
[1] "26-04-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-05-2024"
[1] "04-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-05-2024"
[1] "21-05-2024"
[1] "22-05-2024"
[1] "23-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "24-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-06-2024"
[1] "12-06-2024"
[1] "17-06-2024"
[1] "19-06-2024"
[1] "20-06-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-07-2024"
[1] "12-07-2024"
[1] "22-08-2024"
[1] "26-08-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-09-2024"
[1] "08-10-2024"

data_chiro$Pluie <- res

dd <- unique(data_chiro$Date)
res<- c()
for(d in dd){
  
  if(length(datamap$date[which(datamap$date==d)])>0){
    coord_d <- coord[which(datamap$date==d),]
    delau <- c()
    delau<-tryCatch({ tri2nb(coord_d)}, error=function(e){})
    delau.w<- nb2listw(delau, style="S", zero.policy=TRUE)
    m<-c()
    m <- moran.test(datamap$Temperature[which(datamap$date==d)], delau.w, zero.policy=TRUE)
    if(m$p.value < 0.05){
      spatial_point <- c()
      spatial_point <- SpatialPointsDataFrame(coord_d, as.data.frame(datamap$Temperature[which(datamap$date==d)]))
      geodata <- as.geodata(spatial_point)
      coord <- datamap[,c(4,3)]
      variograme <- variog(geodata, option  = "sm")
      m.d <- max(variograme$u)
      if(length(which(is.na(variograme$v)==T))>0){m.d=variograme$u[which(is.na(variograme$v))][1]}
      variograme <- variog(geodata, option  = "sm",max.dist = m.d)
      mod1 <- variofit(variograme,ini.cov.pars = c((0.9*max(variograme$v,na.rm = T)[1]-variograme$u[1]),variograme$u[which(variograme$v==max(variograme$v,na.rm = T)[1])][1]),nugget=variograme$v[1],"exponential")
      res[which(data_chiro$Date==d)] <- as.numeric(krige.conv(geodata, loc = data_chiro[which(data_chiro$Date==d),c(11,10)] , krige = krige.control(obj.m = mod1))$predict)
      
    }else{
      res[which(data_chiro$Date==d)] <- mean(datamap$Temperature[which(datamap$date==d)])
    }   
    
  }else{
    res[which(data_chiro$Date==d)] <- NA
  }
  print(d)
  
}

[1] "05-02-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-03-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-07-2016"
[1] "13-09-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-09-2016"
[1] "25-10-2016"
[1] "26-10-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-01-2017"
[1] "25-03-2017"
[1] "26-03-2017"
[1] "27-03-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-04-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-04-2017"
[1] "28-05-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-06-2017"
[1] "20-06-2017"
[1] "06-07-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-07-2017"
[1] "01-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-08-2017"
[1] "12-08-2017"
[1] "16-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-09-2017"
[1] "14-09-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-09-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-11-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-11-2017"
[1] "09-12-2017"
[1] "19-01-2018"
[1] "12-02-2018"
[1] "13-02-2018"
[1] "15-02-2018"
[1] "17-02-2018"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-05-2019"
[1] "16-05-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-05-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-05-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-05-2019"
[1] "31-05-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-06-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-06-2019"
[1] "12-06-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-06-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-06-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-06-2019"
[1] "27-06-2019"
[1] "03-07-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-07-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-07-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-07-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-07-2019"
[1] "30-07-2019"
[1] "01-08-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-08-2019"
[1] "14-08-2019"
[1] "18-08-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-08-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-09-2019"
[1] "05-09-2019"
[1] "12-09-2019"
[1] "18-09-2019"
[1] "21-09-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-09-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-10-2019"
[1] "28-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-11-2019"
[1] "04-11-2019"
[1] "05-11-2019"
[1] "06-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "18-11-2019"
[1] "20-11-2019"
[1] "22-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-06-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-10-2020"
[1] "07-11-2020"
[1] "08-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "09-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-01-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-04-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-07-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-10-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-10-2021"
[1] "11-11-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-11-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-12-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-12-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-12-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-02-2022"
[1] "01-03-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-03-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-05-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-06-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-06-2022"
[1] "30-06-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-07-2022"
[1] "09-07-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-07-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "29-08-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-09-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "09-09-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-09-2022"
[1] "19-09-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-10-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-11-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-11-2022"
[1] "05-11-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-11-2022"
[1] "28-01-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "29-01-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-02-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-03-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-03-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-03-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-04-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-04-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-04-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-05-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-05-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "29-05-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-05-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-05-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-06-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-06-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-06-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-06-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-06-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-09-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-09-2023"
[1] "27-09-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-09-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-09-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-10-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-10-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-10-2023"
[1] "13-10-2023"
[1] "14-10-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-10-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-10-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-10-2023"
[1] "28-10-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-10-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-10-2023"
[1] "01-11-2023"
[1] "02-11-2023"
[1] "09-11-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-11-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-12-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-12-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-02-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-02-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-02-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-02-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-04-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-04-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-05-2024"
[1] "04-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "24-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-06-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-06-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-06-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-06-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-06-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-07-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-07-2024"
[1] "22-08-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-08-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-09-2024"
[1] "08-10-2024"

data_chiro$Température <- res

dd <- unique(data_chiro$Date)
res<- c()
for(d in dd){
  
  if(length(datamap$date[which(datamap$date==d)])>0){
    coord_d <- coord[which(datamap$date==d & is.na(datamap$ampli_Temp)==F),]
    delau <- c()
    delau<-tryCatch({ tri2nb(coord_d)}, error=function(e){})
    delau.w <- c()
    delau.w<- nb2listw(delau, style="S", zero.policy=TRUE)
    m<-c()
    m <- moran.test(datamap$ampli_Temp[which(datamap$date==d & is.na(datamap$ampli_Temp)==F)], delau.w, zero.policy=TRUE)
    if(m$p.value < 0.05){
      spatial_point <- c()
      spatial_point <- SpatialPointsDataFrame(coord_d, as.data.frame(datamap$wind_ms[which(datamap$date==d & is.na(datamap$ampli_Temp)==F)]))
      geodata <- as.geodata(spatial_point)
      coord <- datamap[,c(4,3)]
      variograme <- variog(geodata, option  = "sm")
      m.d <- max(variograme$u)
      if(length(which(is.na(variograme$v)==T))>0){m.d=variograme$u[which(is.na(variograme$v))][1]}
      variograme <- variog(geodata, option  = "sm",max.dist = m.d)
      mod1 <- variofit(variograme,ini.cov.pars = c((0.9*max(variograme$v,na.rm = T)[1]-variograme$u[1]),variograme$u[which(variograme$v==max(variograme$v,na.rm = T)[1])][1]),nugget=variograme$v[1],"exponential")
      res[which(data_chiro$Date==d)] <- as.numeric(krige.conv(geodata, loc = data_chiro[which(data_chiro$Date==d),c(11,10)] , krige = krige.control(obj.m = mod1))$predict)
      
    }else{
      res[which(data_chiro$Date==d)] <- mean(datamap$ampli_Temp[which(datamap$date==d)],na.rm = T)
      
    }   
    
  }else{
    res[which(data_chiro$Date==d)] <- NA
  }
  print(d)
  
  
}

[1] "05-02-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-03-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-07-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-09-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-09-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-10-2016"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-10-2016"
[1] "04-01-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-03-2017"
[1] "26-03-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-03-2017"
[1] "16-04-2017"
[1] "17-04-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-05-2017"
as.geodata: 1 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-06-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-06-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-07-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-07-2017"
[1] "01-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-09-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-09-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-09-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-11-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-11-2017"
[1] "09-12-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-01-2018"
[1] "12-02-2018"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-02-2018"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-02-2018"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-02-2018"
as.geodata: 5 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-05-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-05-2019"
[1] "17-05-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-05-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-05-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-05-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-06-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-06-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-06-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-06-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-06-2019"
[1] "21-06-2019"
as.geodata: 6 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-06-2019"
[1] "03-07-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-07-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-07-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-07-2019"
as.geodata: 3 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-07-2019"
[1] "30-07-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-08-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-08-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-08-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "18-08-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-08-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-09-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-09-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-09-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "18-09-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-09-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-09-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-10-2019"
as.geodata: 4 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-10-2019"
as.geodata: 6 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-10-2019"
as.geodata: 6 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-10-2019"
as.geodata: 6 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-10-2019"
as.geodata: 6 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-10-2019"
as.geodata: 6 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-10-2019"
as.geodata: 6 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-10-2019"
as.geodata: 7 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-11-2019"
as.geodata: 7 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-11-2019"
as.geodata: 7 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-11-2019"
as.geodata: 7 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-11-2019"
as.geodata: 7 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-11-2019"
as.geodata: 8 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-11-2019"
as.geodata: 8 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "18-11-2019"
as.geodata: 8 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-11-2019"
as.geodata: 8 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-11-2019"
as.geodata: 8 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-06-2020"
as.geodata: 9 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-10-2020"
as.geodata: 9 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-11-2020"
as.geodata: 9 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-11-2020"
as.geodata: 9 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "09-11-2020"
as.geodata: 9 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-11-2020"
as.geodata: 9 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-11-2020"
as.geodata: 9 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-11-2020"
as.geodata: 9 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-11-2020"
as.geodata: 10 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-01-2021"
as.geodata: 11 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-04-2021"
as.geodata: 12 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-07-2021"
as.geodata: 12 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-10-2021"
as.geodata: 12 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-10-2021"
as.geodata: 13 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-11-2021"
as.geodata: 14 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-11-2021"
as.geodata: 13 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-12-2021"
as.geodata: 13 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-12-2021"
as.geodata: 13 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-12-2021"
as.geodata: 13 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-02-2022"
as.geodata: 13 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-03-2022"
as.geodata: 13 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-03-2022"
as.geodata: 13 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-04-2022"
[1] "04-04-2022"
as.geodata: 14 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-04-2022"
as.geodata: 14 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-04-2022"
as.geodata: 14 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-04-2022"
as.geodata: 15 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-04-2022"
as.geodata: 15 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-05-2022"
as.geodata: 15 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-06-2022"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-06-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-06-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-07-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "09-07-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-07-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-08-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-08-2022"
as.geodata: 14 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-08-2022"
as.geodata: 14 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-08-2022"
as.geodata: 14 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-08-2022"
as.geodata: 14 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-08-2022"
as.geodata: 15 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "29-08-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-09-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "09-09-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-09-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-09-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-10-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-11-2022"
as.geodata: 16 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-11-2022"
as.geodata: 15 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-11-2022"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-11-2022"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-01-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "29-01-2023"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-02-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-03-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-03-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-03-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-04-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-04-2023"
as.geodata: 21 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-04-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-05-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-05-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "29-05-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-05-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-05-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-06-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-06-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-06-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-06-2023"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-06-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-09-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-09-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-09-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-09-2023"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-09-2023"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-10-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-10-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-10-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-10-2023"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "14-10-2023"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-10-2023"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-10-2023"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-10-2023"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-10-2023"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "30-10-2023"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-10-2023"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-11-2023"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-11-2023"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "09-11-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-11-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-12-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-12-2023"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-01-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-01-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-01-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-01-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-01-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-02-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-02-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-02-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-02-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "05-03-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-03-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-03-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-03-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-03-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-03-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "06-04-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-04-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-05-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "03-05-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-05-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-05-2024"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "21-05-2024"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-05-2024"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "23-05-2024"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "24-05-2024"
as.geodata: 17 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "25-05-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-06-2024"
[1] "12-06-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-06-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-06-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-06-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-07-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-07-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-08-2024"
as.geodata: 20 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-08-2024"
as.geodata: 19 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-09-2024"
as.geodata: 18 points removed due to NA in the data
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-10-2024"

data_chiro$Amplitude <- res

dd <- unique(data_chiro$Date)
res<- c()

for(d in dd){
  
  if(length(datamap$date[which(datamap$date==d)])>0){
    coord_d <- coord[which(datamap$date==d & is.na(datamap$wind_ms)==F),]
    delau <- c()
    delau<-tryCatch({ tri2nb(coord_d)}, error=function(e){})
    delau.w <- c()
    delau.w<- nb2listw(delau, style="S", zero.policy=TRUE)
    m<-c()
    m <- moran.test(datamap$wind_ms[which(datamap$date==d & is.na(datamap$wind_ms)==F)], delau.w, zero.policy=TRUE)
    if(m$p.value < 0.05){
      spatial_point <- c()
      spatial_point <- SpatialPointsDataFrame(coord_d, as.data.frame(datamap$wind_ms[which(datamap$date==d & is.na(datamap$wind_ms)==F)]))
      geodata <- as.geodata(spatial_point)
      coord <- datamap[,c(4,3)]
      variograme <- variog(geodata, option  = "sm")
      m.d <- max(variograme$u)
      if(length(which(is.na(variograme$v)==T))>0){m.d=variograme$u[which(is.na(variograme$v))][1]}
      variograme <- variog(geodata, option  = "sm",max.dist = m.d)
      mod1 <- variofit(variograme,ini.cov.pars = c((0.9*max(variograme$v,na.rm = T)[1]-variograme$u[1]),variograme$u[which(variograme$v==max(variograme$v,na.rm = T)[1])][1]),nugget=variograme$v[1],"exponential")
      res[which(data_chiro$Date==d)] <- as.numeric(krige.conv(geodata, loc = data_chiro[which(data_chiro$Date==d),c(11,10)] , krige = krige.control(obj.m = mod1))$predict)
      
    }else{
      res[which(data_chiro$Date==d)] <- mean(datamap$wind_ms[which(datamap$date==d)],na.rm = T)
      
    }   
    
  }else{
    res[which(data_chiro$Date==d)] <- NA
  }
  print(d)
  
  
}

[1] "05-02-2016"
[1] "27-03-2016"
[1] "06-07-2016"
[1] "13-09-2016"
[1] "15-09-2016"
[1] "25-10-2016"
[1] "26-10-2016"
[1] "04-01-2017"
[1] "25-03-2017"
[1] "26-03-2017"
[1] "27-03-2017"
[1] "16-04-2017"
[1] "17-04-2017"
[1] "28-05-2017"
[1] "17-06-2017"
[1] "20-06-2017"
[1] "06-07-2017"
[1] "23-07-2017"
[1] "01-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-08-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-09-2017"
[1] "14-09-2017"
[1] "27-09-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-10-2017"
[1] "30-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "31-10-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "01-11-2017"
[1] "25-11-2017"
[1] "09-12-2017"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-01-2018"
[1] "12-02-2018"
[1] "13-02-2018"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-02-2018"
[1] "17-02-2018"
[1] "10-05-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-05-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-05-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-05-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-05-2019"
[1] "31-05-2019"
[1] "05-06-2019"
[1] "06-06-2019"
[1] "12-06-2019"
[1] "14-06-2019"
[1] "19-06-2019"
[1] "21-06-2019"
[1] "27-06-2019"
[1] "03-07-2019"
[1] "05-07-2019"
[1] "10-07-2019"
[1] "12-07-2019"
[1] "17-07-2019"
[1] "30-07-2019"
[1] "01-08-2019"
[1] "05-08-2019"
[1] "14-08-2019"
[1] "18-08-2019"
[1] "22-08-2019"
[1] "02-09-2019"
[1] "05-09-2019"
[1] "12-09-2019"
[1] "18-09-2019"
[1] "21-09-2019"
[1] "30-09-2019"
[1] "07-10-2019"
[1] "08-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-10-2019"
[1] "21-10-2019"
[1] "25-10-2019"
[1] "26-10-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "27-10-2019"
[1] "28-10-2019"
[1] "01-11-2019"
[1] "03-11-2019"
[1] "04-11-2019"
[1] "05-11-2019"
[1] "06-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "17-11-2019"
[1] "18-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-11-2019"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-11-2019"
[1] "20-06-2020"
[1] "11-10-2020"
[1] "07-11-2020"
[1] "08-11-2020"
[1] "09-11-2020"
[1] "10-11-2020"
[1] "11-11-2020"
[1] "12-11-2020"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "13-11-2020"
[1] "21-01-2021"
[1] "27-04-2021"
[1] "14-07-2021"
[1] "02-10-2021"
[1] "16-10-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "11-11-2021"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "28-11-2021"
[1] "06-12-2021"
[1] "07-12-2021"
[1] "08-12-2021"
[1] "28-02-2022"
[1] "01-03-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-03-2022"
[1] "03-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-04-2022"
[1] "11-04-2022"
[1] "12-04-2022"
[1] "25-04-2022"
[1] "28-04-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-05-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "15-06-2022"
[1] "22-06-2022"
[1] "30-06-2022"
[1] "08-07-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "09-07-2022"
[1] "20-07-2022"
[1] "06-08-2022"
[1] "19-08-2022"
[1] "20-08-2022"
[1] "23-08-2022"
[1] "27-08-2022"
[1] "28-08-2022"
[1] "29-08-2022"
[1] "02-09-2022"
[1] "09-09-2022"
[1] "13-09-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "19-09-2022"
[1] "01-10-2022"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "02-11-2022"
[1] "04-11-2022"
[1] "05-11-2022"
[1] "15-11-2022"
[1] "28-01-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "29-01-2023"
[1] "21-02-2023"
[1] "15-03-2023"
[1] "17-03-2023"
[1] "27-03-2023"
[1] "20-04-2023"
[1] "23-04-2023"
[1] "27-04-2023"
[1] "06-05-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-05-2023"
[1] "29-05-2023"
[1] "30-05-2023"
[1] "31-05-2023"
[1] "01-06-2023"
[1] "03-06-2023"
[1] "26-06-2023"
[1] "27-06-2023"
[1] "28-06-2023"
[1] "08-09-2023"
[1] "14-09-2023"
[1] "27-09-2023"
[1] "28-09-2023"
[1] "30-09-2023"
[1] "01-10-2023"
[1] "02-10-2023"
[1] "07-10-2023"
[1] "13-10-2023"
[1] "14-10-2023"
[1] "20-10-2023"
[1] "26-10-2023"
[1] "27-10-2023"
[1] "28-10-2023"
[1] "30-10-2023"
[1] "31-10-2023"
[1] "01-11-2023"
[1] "02-11-2023"
[1] "09-11-2023"
[1] "23-11-2023"
[1] "19-12-2023"
[1] "20-12-2023"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "10-01-2024"
[1] "12-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "16-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "20-01-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-01-2024"
[1] "07-02-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "08-02-2024"
[1] "15-02-2024"
[1] "21-02-2024"
[1] "05-03-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "12-03-2024"
[1] "13-03-2024"
[1] "21-03-2024"
[1] "28-03-2024"
[1] "31-03-2024"
[1] "06-04-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-04-2024"
[1] "02-05-2024"
[1] "03-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "04-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "07-05-2024"
[1] "21-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "22-05-2024"
[1] "23-05-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "24-05-2024"
[1] "25-05-2024"
[1] "11-06-2024"
[1] "12-06-2024"
[1] "17-06-2024"
[1] "19-06-2024"
[1] "20-06-2024"
[1] "11-07-2024"
[1] "12-07-2024"
[1] "22-08-2024"
variog: computing omnidirectional variogram
variog: computing omnidirectional variogram
variofit: covariance model used is exponential 
variofit: weights used: npairs 
variofit: minimisation function used: optim 
krige.conv: model with constant mean
krige.conv: Kriging performed using global neighbourhood 
[1] "26-08-2024"
[1] "26-09-2024"
[1] "08-10-2024"

data_chiro$Vent <- res

data_chiro <- data_chiro %>%
  mutate(Repro_binary = ifelse(Repro != 0, 1, 0)) 

data_bin <- data_chiro %>%
  group_by(Date) %>%
  summarise(
    total_individuals = n(), 
    repro_positive = sum(Repro_binary), 
    proportion_repro = repro_positive / total_individuals 
  )

data_bin <- data_bin %>%
  mutate(Date = as.Date(Date, format = "%d-%m-%Y"))

data_chiro <- data_chiro %>%
  mutate(Date = as.Date(Date, format = "%d-%m-%Y"))

data_chiro$Lune <-Lune(data_chiro$Date)


data_final <- data_bin %>%
  left_join(data_chiro %>%
              select(julian_day, Year, LAT, LON, Lune, Pluie, Température, Vent, Amplitude, Date) %>%
              distinct(),
            by = "Date")

II. Bayesian approach

We will use the Bayesian approach to estimate the effect of those different covariates on the frequency of reproductive individuals…

A. Quick introduction to the Bayesian approach

This theorem is based on conditional probabilities :

\(P(B \mid A) = \displaystyle{\frac{ P(A \mid B) \; P(B)}{P(A)}}\)

But we might try to explain it more simply as this :

\[P(\text{hypothesis} \mid \text{data}) = \frac{ P(\text{data} \mid \text{hypothesis}) \; P(\text{hypothesis})}{P(\text{data})}\]

B. Little exploration of how our response variable react with different covariates

par(mfrow=c(2,2))

ggplot(data_final, aes(Amplitude, repro_positive)) +
  geom_point() + 
  geom_smooth(method = "loess") +
  scale_y_continuous(trans='log10')

ggplot(data_final, aes(julian_day, repro_positive)) +
  geom_point() + 
  geom_smooth(method = "loess") +
  scale_y_continuous(trans='log10')

ggplot(data_final, aes(Pluie, repro_positive)) +
  geom_point() + 
  geom_smooth(method = "loess")

ggplot(data_final, aes(Température, repro_positive)) +
  geom_point() + 
  geom_smooth(method = "loess") +
  scale_y_continuous(trans='log10')

At this point, we might be tempted to see whether or not these climatic parameters influence the proportion of reproductive individuals… let’s try!!

C. Study of climatic influence on the proportion of breeding indidivids

Let’s see the distribution of reproductive individuals :

hist(data_final$repro_positive, breaks = 20)

The number of individuals showing reproductive activity seems to follow a negative-binomial distribution, but we encountered some issues to write and fit model using a negative binomial distribution… anyway ! Let’s try to investigate with a binomial !

First, we create an object containing our covariates of interest, we use a common practice which consists of centering-reducing the values of covariates:

datax <- list(
  N = nrow(data_final), 
  repro_positive = data_final$repro_positive,  
  total_individuals = data_final$total_individuals,  
  temp = (data_final$Température - mean(data_final$Température)) / sd(data_final$Température),  
  rain = (data_final$Pluie - mean(data_final$Pluie)) / sd(data_final$Pluie),
  julian_day = (data_final$julian_day - mean(data_final$julian_day)) / sd(data_final$julian_day))

We write our null model and we assume that the number of reproductive individuals follows a binomial distribution. Specifically, we define our model as follows:

\[ \text{repro_positive}[i] \sim \text{Binomial}(p[i], \text{total_individuals}[i]) \]

In this model, \(\text{repro_positive}[i]\) represents the number of reproductive individuals observed for each observation ( i ), while \(\text{total_individuals}[i]\) is the total number of individuals in the corresponding group. The parameter \(p[i]\) is the probability of an individual being reproductive, modeled using the logit link function, where ( a ) is a parameter drawn from a normal distribution with mean 0 and a large variance (0.001), reflecting our prior belief about its value. \[ \text{logit}(p[i]) = a \]

Here is the null model in Jags language:

model_null <- 
  paste("
model {
  for (i in 1:N) {
    repro_positive[i] ~ dbin(p[i], total_individuals[i])  
    logit(p[i]) <- a
  }
  a ~ dnorm(0, 0.001)  
}
")

Then we set the initial values for the Monte-Carlo Markov Chains, we choose to set two chains to assess initial at different values to see if both converge to same distribution.

init1 <- list (a = -0.5)
init2 <- list (a = 0.5)   
inits <- list(init1,init2)

We specify the parameter we want to estimate :

parameters <- c("a")

The run the model with the function jags() :

chiro_model_null <- jags(data = datax,
              inits = inits, 
              parameters.to.save = parameters, 
              model.file = textConnection(model_null), 
              n.chains = 2, 
              n.iter = 5000, 
              n.burnin = 1000)

Compiling model graph
   Resolving undeclared variables
   Allocating nodes
Graph information:
   Observed stochastic nodes: 246
   Unobserved stochastic nodes: 1
   Total graph size: 497

Initializing model

After fitting the model, we can explore how it went through some useful tools :

traceplot() allows to see how MCMC converge

traceplot(chiro_model_null, mfrow = c(1,1), varnames = c("a"), ask = F)

autocorr.plot(as.mcmc(chiro_model_null), ask = F)

We can also print the object that contain the model to check other features like the estimate of the parameter, the deviance, credible intervals or the number of values used to assess the posterior distribution (n.eff)

print(chiro_model_null)

Inference for Bugs model at "4", fit using jags,
 2 chains, each with 5000 iterations (first 1000 discarded), n.thin = 4
 n.sims = 2000 iterations saved
         mu.vect sd.vect    2.5%     25%     50%     75%   97.5%  Rhat n.eff
a          1.655   0.073   1.517   1.605   1.655   1.704   1.802 1.002   870
deviance 755.015   1.416 754.025 754.122 754.475 755.305 759.055 1.001  2000

For each parameter, n.eff is a crude measure of effective sample size,
and Rhat is the potential scale reduction factor (at convergence, Rhat=1).

DIC info (using the rule, pD = var(deviance)/2)
pD = 1.0 and DIC = 756.0
DIC is an estimate of expected predictive error (lower deviance is better).

Now, we put the estimated values of the two MCMC chains in the same object :

res<-as.mcmc(chiro_model_null)
res<-rbind(res[[1]], res[[2]])

Then we can compute the mean of the estimated values of the intercept that are < 0 :

mean(res[,'a']<0)

[1] 0

And we can plot the estimated posterior distribution of our parameter ‘a’ :

par(mfrow=c(1,1))
plot(density(res[,'a']),xlab="",ylab="", main="intercept",lwd=3,xlim=c(-0.1,2))
abline(v=0,col='red',lwd=2)

Remember that we defined a non-informative prior on parameter ‘a’ : a ~ dnorm(0, 0.001) to see if test the effect of ‘a’ on \(p[i]\)

You can compute the Watanabe-Akaike Information Criterion :

samples_waic_null<-jags.samples(chiro_model_null$model, c("WAIC", "deviance"), type = "mean",
                           n.iter = 10000,
                           n.burnin = 1000,
                           n.thin = 1)
samples_waic_null$p_waic<-samples_waic_null$WAIC
samples_waic_null$waic<-samples_waic_null$deviance + samples_waic_null$p_waic
tmp<-sapply(samples_waic_null, sum)
waic_null<-round(c(waic = tmp[["waic"]], p_waic = tmp[["p_waic"]]),1)
waic_null

  waic p_waic 
 758.2    3.1

We write and fit several models with different parameters (rain, temperature, julian day and thermal amplitude) and we compute the WAIC for each of them:

Here the table of the different models we fitted :

waic_table<-read.table("waic_model.txt", header = T)
waic_table

                          model  waic p_waic
1          model_rain_temp_jday 523.3   18.4
2 model_rain_temp_jday_thermamp 528.2   22.6
3                    model_jday 533.9   11.9
4               model_rain_jday 541.1   17.6
5               model_rain_temp 594.0   10.8
6                    model_null 616.3    3.2
7                    model_rain 619.9    5.7

D. The best model

So our best model is the one with rain/temperature/julian_day covariates, there it is :

model_rain_temp_jday <- 
  paste("
model {
  for (i in 1:N) {
    repro_positive[i] ~ dbin(p[i], total_individuals[i])  # Distribution binomiale
    logit(p[i]) <- a + b.rain * rain[i] 
                     + b.rain2 * pow(rain[i],2)  
                     + b.temp * temp[i] 
                     + b.julian_day * julian_day[i]
                     + b.julian_day2 * pow(julian_day[i], 2)
                     + b.julian_day3 * pow(julian_day[i], 3)
                     + b.julian_day4 * pow(julian_day[i], 4)
  }
  
  a ~ dnorm(0, 0.001)  # Prior sur l'intercept
  b.rain ~ dnorm(0, 0.001)
  b.rain2 ~ dnorm(0, 0.001)
  b.temp ~ dnorm(0, 0.001)
  b.julian_day ~ dnorm(0, 0.001)
  b.julian_day2 ~ dnorm(0, 0.001) 
  b.julian_day3 ~ dnorm(0, 0.001)
  b.julian_day4 ~ dnorm(0, 0.001)
}
")

We decide to assess polynomial effect on \(rain^2\) and on \(julian\_day^4\) because of the non-linearity effect on \(repro\_positive\)

And as we made before, we set the initial values for both MCMC, we specify the parameters to be estimated :

init1 <- list (a = -0.5, 
               b.rain = -0.5, 
               b.rain2 = -0.5, 
               b.temp = -0.5,
               b.julian_day = -0.5,
               b.julian_day2= -0.5,
               b.julian_day3= -0.5,
               b.julian_day4 = -0.5) 

init2 <- list (a = 0.5, 
               b.rain = 0.5, 
               b.rain2 = 0.5, 
               b.temp = 0.5, 
               b.julian_day = 0.5,
               b.julian_day2= 0.5,
               b.julian_day3= 0.5,
               b.julian_day4 = 0.5)

inits <- list(init1,init2)

parameters <- c("a",
                "b.rain", 
                "b.rain2", 
                "b.temp",
                "b.julian_day" ,
                "b.julian_day2",
                "b.julian_day3",
                "b.julian_day4")

And we run jags() :

chiro_rain_temp_jday <- jags(data = datax,
                        inits = inits, 
                        parameters.to.save = parameters, 
                        model.file = textConnection(model_rain_temp_jday), 
                        n.chains = 2, 
                        n.iter = 10000, 
                        n.burnin = 1000)

Compiling model graph
   Resolving undeclared variables
   Allocating nodes
Graph information:
   Observed stochastic nodes: 246
   Unobserved stochastic nodes: 8
   Total graph size: 2709

Initializing model

chiro_rain_temp_jday

Inference for Bugs model at "5", fit using jags,
 2 chains, each with 10000 iterations (first 1000 discarded), n.thin = 9
 n.sims = 2000 iterations saved
              mu.vect sd.vect    2.5%     25%     50%     75%   97.5%  Rhat
a               0.421   0.152   0.138   0.318   0.421   0.523   0.711 1.001
b.julian_day    0.152   0.244  -0.344  -0.009   0.153   0.312   0.633 1.001
b.julian_day2   2.317   0.324   1.672   2.109   2.321   2.536   2.933 1.001
b.julian_day3  -0.216   0.150  -0.502  -0.316  -0.219  -0.119   0.089 1.003
b.julian_day4  -0.572   0.130  -0.815  -0.657  -0.575  -0.492  -0.308 1.001
b.rain          0.437   0.218   0.015   0.292   0.433   0.583   0.857 1.001
b.rain2         0.058   0.103  -0.075  -0.017   0.033   0.110   0.310 1.001
b.temp          0.564   0.129   0.300   0.479   0.563   0.645   0.820 1.001
deviance      616.997   7.763 610.779 613.810 616.213 619.039 626.611 1.129
              n.eff
a              2000
b.julian_day   2000
b.julian_day2  2000
b.julian_day3  2000
b.julian_day4  2000
b.rain         2000
b.rain2        2000
b.temp         2000
deviance       2000

For each parameter, n.eff is a crude measure of effective sample size,
and Rhat is the potential scale reduction factor (at convergence, Rhat=1).

DIC info (using the rule, pD = var(deviance)/2)
pD = 30.1 and DIC = 647.1
DIC is an estimate of expected predictive error (lower deviance is better).

As we made before, we check how the MCMC converged and we trace the autocorrelation plot :

traceplot(chiro_rain_temp_jday, mfrow = c(2,4), varnames = c("a","b.rain", "b.rain2","b.temp","b.julian_day" ,"b.julian_day2","b.julian_day3","b.julian_day4"), ask = F)

autocorr.plot(as.mcmc(chiro_rain_temp_jday), ask = F)

We put the estimate values in the same object :

res<-as.mcmc(chiro_rain_temp_jday)
res<-rbind(res[[1]], res[[2]])

And we plot the estimated distribution of every parameters :

par(mfrow=c(2,4))
plot(density(res[,'a']),xlab="",ylab="", main="Intercept",lwd=3,xlim=c(-0.1,2))
abline(v=0,col='red',lwd=2)
plot(density(res[,'b.rain']),xlab="",ylab="", main="Rainfall",lwd=3)
abline(v=0,col='red',lwd=2)
plot(density(res[,'b.rain2']),xlab="",ylab="", main="Rainfall",lwd=3)
abline(v=0,col='red',lwd=2)
plot(density(res[,'b.temp']),xlab="",ylab="", main="Temperature",lwd=3, xlim=c(-0.1,1.5))
abline(v=0,col='red',lwd=2)
plot(density(res[,'b.julian_day']),xlab="",ylab="", main="julian_day",lwd=3)
abline(v=0,col='red',lwd=2)
plot(density(res[,'b.julian_day2']),xlab="",ylab="", main="julian_day2",lwd=3, xlim = c(0,4))
abline(v=0,col='red',lwd=2)
plot(density(res[,'b.julian_day3']),xlab="",ylab="", main="julian_day3",lwd=3)
abline(v=0,col='red',lwd=2)
plot(density(res[,'b.julian_day4']),xlab="",ylab="", main="julian_day4",lwd=3)
abline(v=0,col='red',lwd=2)

E. Using our model

Now let’s try to predict some values from our model to better understand the different relations between \(repro\_positive\) and the different covariates.

First, let’s simulate some data !

pred_data <- expand.grid(
  rain = seq(min(datax$rain), max(datax$rain), length.out = 1000), 
  temp = median(datax$temp),
  julian_day = median(datax$julian_day))

This grid will allow us to simulate data from the source data.frame “data_final”, for some reasons that we don’t clearly understand… we need to sample each covariate at once and fix the other one to the median.

Then we compute \(p[i]\) for each estimate value from the model : \[ \text{logit}(p_i) = a + b_{\text{rain}} \cdot \text{rain}_i + b_{\text{rain2}} \cdot \text{rain}_i^2 + b_{\text{temp}} \cdot \text{temp}_i + b_{\text{julian_day}} \cdot \text{julian_day}_i + b_{\text{julian_day2}} \cdot \text{julian_day}_i^2 + b_{\text{julian_day3}} \cdot \text{julian_day}_i^3 + b_{\text{julian_day4}} \cdot \text{julian_day}_i^4 \]

n_sim<-1000

p_sim <- matrix(NA, nrow = n_sim, ncol = nrow(pred_data))

for (i in 1:n_sim) {
  sample_idx <- sample(1:nrow(res), 1)
  a <- res[sample_idx, "a"]
  b.rain <- res[sample_idx, "b.rain"]
  b.rain2 <- res[sample_idx, "b.rain2"]
  b.temp <- res[sample_idx, "b.temp"]
  b.julian_day <- res[sample_idx, "b.julian_day"]
  b.julian_day2 <- res[sample_idx, "b.julian_day2"]
  b.julian_day3 <- res[sample_idx, "b.julian_day3"]
  b.julian_day4 <- res[sample_idx, "b.julian_day4"]
  
  p_sim[i, ] <- plogis(
    a + 
      b.rain * pred_data$rain + b.rain2 * pred_data$rain^2 +
      b.temp * pred_data$temp +
      b.julian_day * pred_data$julian_day + 
      b.julian_day2 * pred_data$julian_day^2 +
      b.julian_day3 * pred_data$julian_day^3 + 
      b.julian_day4 * pred_data$julian_day^4
  )
}

Then we compute our credible intervals :

pred_data <- pred_data %>%
  mutate(
    mean_p = apply(p_sim, 2, mean),
    lower = apply(p_sim, 2, quantile, probs = 0.025),
    upper = apply(p_sim, 2, quantile, probs = 0.975)
  )

We want to get back to the natural scale for the X-axis :

mean(data_final$Pluie)

[1] 6.95171

sd(data_final$Pluie)

[1] 10.36198

pred_data$rain <- (pred_data$rain * 10.71609) + 7.087872

Then we can plot our response variable with a covariate ! here the rain :

ggplot(pred_data, aes(x = rain)) +
  geom_line(aes(y = mean_p), color = "blue") +
  geom_ribbon(aes(ymin = lower, ymax = upper), alpha = 0.2) +
  labs(x = "Rain (Natural Scale)", y = "Probability of Reproduction") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))+
  scale_y_continuous(limits = c(0, 1))

Same task for julian_day… yes it’s a bit boring, we need to improve the code !

pred_data <- expand.grid(
  rain = median(datax$temp), 
  temp = median(datax$temp),
  julian_day = seq(min(datax$julian_day), max(datax$julian_day), length.out = 1000))
p_sim <- matrix(NA, nrow = n_sim, ncol = nrow(pred_data))

for (i in 1:n_sim) {
  sample_idx <- sample(1:nrow(res), 1)
  a <- res[sample_idx, "a"]
  b.rain <- res[sample_idx, "b.rain"]
  b.rain2 <- res[sample_idx, "b.rain2"]
  b.temp <- res[sample_idx, "b.temp"]
  b.julian_day <- res[sample_idx, "b.julian_day"]
  b.julian_day2 <- res[sample_idx, "b.julian_day2"]
  b.julian_day3 <- res[sample_idx, "b.julian_day3"]
  b.julian_day4 <- res[sample_idx, "b.julian_day4"]
  
  p_sim[i, ] <- plogis(
    a + 
      b.rain * pred_data$rain + b.rain2 * pred_data$rain^2 +
      b.temp * pred_data$temp +
      b.julian_day * pred_data$julian_day + 
      b.julian_day2 * pred_data$julian_day^2 +
      b.julian_day3 * pred_data$julian_day^3 + 
      b.julian_day4 * pred_data$julian_day^4
  )
}
pred_data <- pred_data %>%
  mutate(
    mean_p = apply(p_sim, 2, mean),
    lower = apply(p_sim, 2, quantile, probs = 0.025),
    upper = apply(p_sim, 2, quantile, probs = 0.975)
  )

mean(data_final$julian_day)

[1] 201.5163

sd(data_final$julian_day)

[1] 96.75666

pred_data$julian_day <- (pred_data$julian_day * 95.50893) + 203.0329

ggplot(pred_data, aes(x = julian_day)) +
  geom_line(aes(y = mean_p), color = "blue") +
  geom_ribbon(aes(ymin = lower, ymax = upper), alpha = 0.2) +
  labs(x = "Julian day (Natural Scale)", y = "Probability of Reproduction") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))+
  scale_y_continuous(limits = c(0, 1))

Results: Those results suggest that the probability of being reproductive could be mainly influenced by the two rainy seasons that occurs in the lowlands of French Guiana and, therefore, with the fructification period of many plants. As the extreme majority of the species caugth with mist-nets are frugivorous.

(We would greatly appreciate advice on how to improve this analysis!)

References

Charles-Dominique Pierre, Brosset André, and Jouard Sylvie. 2001. Les Chauves-Souris de Guyane. 49. Muséum national d’histoire naturelle.

Hollowell, Tom, and Robert P Reynolds. 2005. “Checklist of the Terrestrial Vertebrates of the Guiana Shield,” no. 13.

MacSwiney G., M. Cristina, Frank M. Clarke, and Paul A. Racey. 2008. “What You See Is Not What You Get: The Role of Ultrasonic Detectors in Increasing Inventory Completeness in Neotropical Bat Assemblages.” Journal of Applied Ecology 45 (5): 1364–71. https://doi.org/10.1111/j.1365-2664.2008.01531.x.

Ruiz-Ramoni, Damián, Paolo Ramoni-Perazzi, and Mariana Munoz-Romo. 2016. “Reproductive Pattern of the Large Fruit-Eating Bat, Artibeus Amplus (Chiroptera: Phyllostomidae) in the Venezuelan Andes.” Revista de Biología Tropical 65 (1): 335. https://doi.org/10.15517/rbt.v65i1.23863.

Spinu, Vitalie, Garrett Grolemund, and Hadley Wickham. 2010. “Lubridate: Make Dealing with Dates a Little Easier.” https://doi.org/10.32614/CRAN.package.lubridate.

Wickham, Hadley, Romain François, Lionel Henry, Kirill Müller, Davis Vaughan, Posit Software, and PBC. 2023. “Dplyr: A Grammar of Data Manipulation.” https://cran.r-project.org/web/packages/dplyr/index.html.