The main propose of this package is the capability of run time-specific ecological niche models.
We start with a simple example to show the basic functions of the package. We will work with a dataset of Abronia graminea, an endemic lizard from the Mexican Sierra Madre Oriental.
First, we load the tenm R package.
Now we load the abronia dataset, which contains
geographical information about the presence of Abronia graminea
in its area of distribution. This dataset has also information about the
year of observation and the GBIF
doi.
data("abronia")
head(abronia)
#>            species decimalLongitude decimalLatitude year
#> 1 Abronia graminea        -98.17773        19.96523 2014
#> 2 Abronia graminea        -98.13753        19.87006 2014
#> 3 Abronia graminea        -98.07042        19.89668 2014
#> 4 Abronia graminea        -98.13003        19.86861 2014
#> 5 Abronia graminea        -98.14894        19.84450 2014
#> 6 Abronia graminea        -98.15909        19.86878 2014
#>                             gbif_doi
#> 1 https://doi.org/10.15468/dl.teyjm9
#> 2 https://doi.org/10.15468/dl.teyjm9
#> 3 https://doi.org/10.15468/dl.teyjm9
#> 4 https://doi.org/10.15468/dl.teyjm9
#> 5 https://doi.org/10.15468/dl.teyjm9
#> 6 https://doi.org/10.15468/dl.teyjm9
dim(abronia)
#> [1] 106   5We plot the geographic information to see how Abronia graminea is distributed.
Fig. 1. Occurrence points of Abronia graminea. Colors represent the year of observation.
Note that some occurrences are overlapped but belong to different years.
A relevant step when curating occurrence data is to eliminate
duplicated geographical information, which depends on several factors,
including spatial autocorrelation and the spatial resolution of the
modeling layers. Let’s see what happens when we eliminate duplicated
information as defined by the spatial resolution of our modeling layers.
To do this, we will use the tenm::clean_dup function of the
tenm R package.
# Load a modeling layer 
tempora_layers_dir <- system.file("extdata/bio",package = "tenm")
tenm_mask <- terra::rast(file.path(tempora_layers_dir,"1939/bio_01.tif"))
ab_1 <- tenm::clean_dup(data =abronia,
                        longitude = "decimalLongitude",
                        latitude = "decimalLatitude",
                        threshold = terra::res(tenm_mask),
                        by_mask = FALSE,
                        raster_mask = NULL)
tidyr::as_tibble(ab_1)
#> # A tibble: 10 × 5
#>    species          decimalLongitude decimalLatitude  year gbif_doi             
#>    <chr>                       <dbl>           <dbl> <int> <chr>                
#>  1 Abronia graminea            -97.5            19.5  1995 https://doi.org/10.1…
#>  2 Abronia graminea            -97.0            18.2  1993 https://doi.org/10.1…
#>  3 Abronia graminea            -98.0            19.8  1980 https://doi.org/10.1…
#>  4 Abronia graminea            -97.7            19.6  2012 https://doi.org/10.1…
#>  5 Abronia graminea            -97.9            20.1  2015 https://doi.org/10.1…
#>  6 Abronia graminea            -97.4            18.5  1952 https://doi.org/10.1…
#>  7 Abronia graminea            -97.1            18.9  1998 https://doi.org/10.1…
#>  8 Abronia graminea            -97.3            19.0  1983 https://doi.org/10.1…
#>  9 Abronia graminea            -97.3            18.7  1973 https://doi.org/10.1…
#> 10 Abronia graminea            -97.0            19.7  1972 https://doi.org/10.1…After applying our spatial thinning, we obtained only ten
observations from 106 occurrences. We lost a lot of information!!! The
function tenm::clean_dup has a method to clean duplicated
records according to a raster mask layer. The above avoids losing
records that might occur in different pixels but fall within the
distance used as threshold for cleaning.
ab_by_mask <- tenm::clean_dup(data =abronia,
                              longitude = "decimalLongitude",
                              latitude = "decimalLatitude",
                              threshold = terra::res(tenm_mask),
                              by_mask = TRUE,
                              raster_mask = tenm_mask)
tidyr::as_tibble(ab_by_mask)
#> # A tibble: 16 × 5
#>    species          decimalLongitude decimalLatitude  year gbif_doi             
#>    <chr>                       <dbl>           <dbl> <int> <chr>                
#>  1 Abronia graminea            -98.2            20.0  2014 https://doi.org/10.1…
#>  2 Abronia graminea            -98.1            19.9  2014 https://doi.org/10.1…
#>  3 Abronia graminea            -98.1            19.8  2014 https://doi.org/10.1…
#>  4 Abronia graminea            -97.9            19.9  2014 https://doi.org/10.1…
#>  5 Abronia graminea            -97.3            18.7  1963 https://doi.org/10.1…
#>  6 Abronia graminea            -97.1            18.3  1996 https://doi.org/10.1…
#>  7 Abronia graminea            -97.4            18.8  1941 https://doi.org/10.1…
#>  8 Abronia graminea            -97.4            18.7  1988 https://doi.org/10.1…
#>  9 Abronia graminea            -97.0            19.6  1991 https://doi.org/10.1…
#> 10 Abronia graminea            -97.4            19.1  2002 https://doi.org/10.1…
#> 11 Abronia graminea            -97.5            19.5  1995 https://doi.org/10.1…
#> 12 Abronia graminea            -97.0            18.2  1993 https://doi.org/10.1…
#> 13 Abronia graminea            -97.7            19.6  2012 https://doi.org/10.1…
#> 14 Abronia graminea            -97.9            20.1  2015 https://doi.org/10.1…
#> 15 Abronia graminea            -97.1            18.9  1998 https://doi.org/10.1…
#> 16 Abronia graminea            -97.3            19.0  1983 https://doi.org/10.1…We recovered 6 records, not bad! On the other hand, we did not
account for the fact that some occurrences come from different years.
The tenm package is designed to deal with occurrences
coming from different periods as long as the user has environmental
layers matching the years of occurrence observations.
Let’s apply the functions and methods to work with time-specific niche models. First, we load our data.
Now, we indicate the path where our time-specific modeling layers are located.
tempora_layers_dir <- system.file("extdata/bio",package = "tenm")
print(tempora_layers_dir)
#> [1] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio"We explore the structure of the directory that contains our modeling layers.
list.dirs(tempora_layers_dir,recursive = FALSE)
#>  [1] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1939"
#>  [2] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1940"
#>  [3] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1941"
#>  [4] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1950"
#>  [5] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1952"
#>  [6] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1963"
#>  [7] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1969"
#>  [8] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1970"
#>  [9] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1971"
#> [10] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1972"
#> [11] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1973"
#> [12] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1974"
#> [13] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1976"
#> [14] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1977"
#> [15] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1980"
#> [16] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1981"
#> [17] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1982"
#> [18] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1983"
#> [19] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1988"
#> [20] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1991"
#> [21] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1993"
#> [22] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1994"
#> [23] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1995"
#> [24] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1996"
#> [25] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1998"
#> [26] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/2002"
#> [27] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/2008"
#> [28] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/2011"
#> [29] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/2012"
#> [30] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/2014"
#> [31] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/2015"
#> [32] "/Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/2016"Note that the directory contains other directories named with the dates of the modeling layers. Now, we explore some of these dated directories.
# Directory for year 1939
list.files(list.dirs(tempora_layers_dir,
                     recursive = FALSE)[1],
           pattern = ".tif$")
#>  [1] "bio_01.tif" "bio_02.tif" "bio_03.tif" "bio_04.tif" "bio_05.tif"
#>  [6] "bio_06.tif" "bio_07.tif" "bio_08.tif" "bio_09.tif" "bio_10.tif"
#> [11] "bio_11.tif" "bio_12.tif" "bio_13.tif" "bio_14.tif" "bio_15.tif"
#> [16] "bio_16.tif" "bio_17.tif" "bio_18.tif" "bio_19.tif"
# Directory for year 1972
list.files(list.dirs(tempora_layers_dir,
                     recursive = FALSE)[10],
           pattern = ".tif$")
#>  [1] "bio_01.tif" "bio_02.tif" "bio_03.tif" "bio_04.tif" "bio_05.tif"
#>  [6] "bio_06.tif" "bio_07.tif" "bio_08.tif" "bio_09.tif" "bio_10.tif"
#> [11] "bio_11.tif" "bio_12.tif" "bio_13.tif" "bio_14.tif" "bio_15.tif"
#> [16] "bio_16.tif" "bio_17.tif" "bio_18.tif" "bio_19.tif"
# Directory for year 2014
list.files(list.dirs(tempora_layers_dir,
                     recursive = FALSE)[30],
           pattern = ".tif$")
#>  [1] "bio_01.tif" "bio_02.tif" "bio_03.tif" "bio_04.tif" "bio_05.tif"
#>  [6] "bio_06.tif" "bio_07.tif" "bio_08.tif" "bio_09.tif" "bio_10.tif"
#> [11] "bio_11.tif" "bio_12.tif" "bio_13.tif" "bio_14.tif" "bio_15.tif"
#> [16] "bio_16.tif" "bio_17.tif" "bio_18.tif" "bio_19.tif"Note that all dated directories must have the same environmental information. In this example, we used the bioclimatic layers derived from the CHELSAcruts database.
In the following lines of code, we will use a special function of the
tenm R package that will allow us to work with
time-specific data.
data("abronia")
tempora_layers_dir <- system.file("extdata/bio",package = "tenm")
abt <- tenm::sp_temporal_data(occs = abronia,
                              longitude = "decimalLongitude",
                              latitude = "decimalLatitude",
                              sp_date_var = "year",
                              occ_date_format="y",
                              layers_date_format= "y",
                              layers_by_date_dir = tempora_layers_dir,
                              layers_ext="*.tif$")The function tenm::sp_temporal_data is parametrized with
the occurrence dated database. To parametrize the function, we need to
specify the name of the columns that contain the longitude and latitude
data, the column that represents the year of observation, the format of
dates (here years, but see the help of the function for other date
formats), the layers date format, the directory that contains the
time-specific modeling layers and the raster layer extension.
The object abt is a special class called
sp.temporal.modeling that deals with time-specific
information.
In the following line of code, we explore the slots of
abt object.
# See the names of the slots
names(abt)
#> [1] "temporal_df"  "sp_date_var"  "lon_lat_vars" "layers_ext"The abt object has four slots: temporal data.frame
(“temporal_df”), a character vector indicating the date variable
(“sp_date_var”), a character vector showing the names of longitude and
latitude data (“lon_lat_vars”) and another character vector with the
extension of the modeling layers.
Now, we explore the temporal_df slot, which is a data.frame with five columns: longitude, latitude, the time variable (here year), the layer dates, and layers path (the path the temporal niche layers are located).
# See the temporal data.frame
tidyr::as_tibble(head(abt$temporal_df))
#> # A tibble: 6 × 5
#>   decimalLongitude decimalLatitude  year layer_dates layers_path                
#>              <dbl>           <dbl> <int> <date>      <chr>                      
#> 1            -98.2            20.0  2014 2014-01-01  /Library/Frameworks/R.fram…
#> 2            -98.1            19.9  2014 2014-01-01  /Library/Frameworks/R.fram…
#> 3            -98.1            19.9  2014 2014-01-01  /Library/Frameworks/R.fram…
#> 4            -98.1            19.9  2014 2014-01-01  /Library/Frameworks/R.fram…
#> 5            -98.1            19.8  2014 2014-01-01  /Library/Frameworks/R.fram…
#> 6            -98.2            19.9  2014 2014-01-01  /Library/Frameworks/R.fram…As a first step, we will curate our time-specific database using the
function tenm::clean_dup_by_date. This function is
parametrized as the tenm::clean_dup function with the
difference that it thins the data considering the time variable (some
occurrences might be spatially duplicated but belong to other dates, so
in a time-specific context, they are not duplicates).
# Clean duplicates using a raster mask
abtc <- tenm::clean_dup_by_date(this_species = abt,
                                by_mask = TRUE,
                                threshold = terra::res(tenm_mask)[1],
                                raster_mask = tenm_mask[1],
                                n_ngbs = 0)
# Check number of records
head(tidyr::as_tibble(abtc$temporal_df))
#> # A tibble: 6 × 5
#>   decimalLongitude decimalLatitude  year layer_dates layers_path                
#>              <dbl>           <dbl> <int> <date>      <chr>                      
#> 1            -97.3            18.7  1939 1939-01-01  /Library/Frameworks/R.fram…
#> 2            -97.3            18.7  1940 1940-01-01  /Library/Frameworks/R.fram…
#> 3            -97.0            19.6  1941 1941-01-01  /Library/Frameworks/R.fram…
#> 4            -97.3            18.7  1941 1941-01-01  /Library/Frameworks/R.fram…
#> 5            -97.3            18.7  1950 1950-01-01  /Library/Frameworks/R.fram…
#> 6            -97.1            19.7  1950 1950-01-01  /Library/Frameworks/R.fram…
nrow(abtc$temporal_df)
#> [1] 40An improvement of this methodology is that we recover a lot of information. From 10 records thinned using the standard data cleaning process, now we have 40 records; 30 more observations!!! which will allow us to fit more informative models. Let’s compare occurrences from the standard data cleaning process and the time-specific thinning process.
colors1 <- hcl.colors(length(unique(ab_1$year)))
par(mar=c(4,4,2,2),mfrow=c(1,2))
plot(ab_1$decimalLongitude, ab_1$decimalLatitude,
     col=colors1,pch=19, cex=0.75,
     xlab="Longitude",ylab="Latitude",xlim=c(-98.35,-96.7))
legend("bottomleft",legend = sort(unique(ab_1$year))[1:10],
       cex=0.85,pt.cex = 1,bty = "n",
       pch=19,col =colors1[1:10])
colors2 <- hcl.colors(length(unique(abtc$temporal_df$year)))
plot(abtc$temporal_df$decimalLongitude, abtc$temporal_df$decimalLatitude,
     col=colors2,pch=19, cex=0.75,
     xlab="Longitude",ylab="Latitude",xlim=c(-98.35,-96.7))
legend("bottomleft",legend = sort(unique(abtc$temporal_df$year))[1:16],
       cex=0.85,pt.cex = 1,bty = "n",
       pch=19,col =colors2[1:16])
legend("bottomright",
       legend = sort(unique(abronia$year))[17:length(unique(abtc$temporal_df$year))],
       cex=0.85,pt.cex = 1,bty = "n",
       pch=19,col =colors2[17:length(unique(abtc$temporal_df$year))])Fig. 2. Comparison of the spatial distribution of occurrence records for the standard thinning processs and the time-specific thinning process. Left panel shows the records after the standard thinning process. Right panel shows the spatial distribution of the records after the time-specific thinning process; note that some records overlap but are from different years.
After the spatial thinning process, we need to extract environmental
information from our occurrence points. The tenm package
does this using the function tenm::ex_by_date. This
function can be run in parallel by evoking functions of the
future package. To parametrize the function, we need to
specify the “sp.temporal.modeling” object (obtained using the function
tenm::sp_temporal_data or the one from
tenm::clean_dup_by_date) and the proportion of occurrences
to be used as the training dataset. The tenm package uses a
random partition to divide the database into train and test
datasets.
future::plan("multisession",workers=2)
abex <- tenm::ex_by_date(this_species = abtc,
                         train_prop=0.7)
future::plan("sequential")Now, we explore the slot “temporal_df”.
head(abex$temporal_df)
#> # A tibble: 6 × 26
#>   decimalLongitude decimalLatitude  year layer_dates layers_path   cell_ids_year
#>              <dbl>           <dbl> <int> <date>      <chr>                 <dbl>
#> 1            -97.3            18.7  1939 1939-01-01  /Library/Fra…           272
#> 2            -97.3            18.7  1940 1940-01-01  /Library/Fra…           272
#> 3            -97.0            19.6  1941 1941-01-01  /Library/Fra…           173
#> 4            -97.3            18.7  1941 1941-01-01  /Library/Fra…           271
#> 5            -97.3            18.7  1950 1950-01-01  /Library/Fra…           272
#> 6            -97.1            19.7  1950 1950-01-01  /Library/Fra…           173
#> # ℹ 20 more variables: bio_01 <int>, bio_02 <int>, bio_03 <int>, bio_04 <int>,
#> #   bio_05 <int>, bio_06 <int>, bio_07 <int>, bio_08 <int>, bio_09 <int>,
#> #   bio_10 <int>, bio_11 <int>, bio_12 <int>, bio_13 <int>, bio_14 <int>,
#> #   bio_15 <int>, bio_16 <int>, bio_17 <int>, bio_18 <int>, bio_19 <int>,
#> #   trian_test <chr>It has 40 rows and columns with the time-specific environmental values and an additional column indicating if the observation will be used as train or test.
The tenm package uses environmental background to
compute the ROC and partial ROC test and estimate the prevalence of the
species in the environmental space (proportion of environmental points
inside the niche model). We will generate 10,000 environmental
background points using as calibration area and a neighborhood of 10
pixels around each occurrence point (buffer_ngbs parameter).
future::plan("multisession",workers=2)
abbg <- tenm::bg_by_date(this_species = abex,
                         buffer_ngbs=10,n_bg=10000)
future::plan("sequential")
head(abbg$env_bg)
#>                                                                                      ID_YEAR
#> 1 /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1939
#> 2 /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1939
#> 3 /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1939
#> 4 /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1939
#> 5 /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1939
#> 6 /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library/tenm/extdata/bio/1939
#>   decimalLongitude decimalLatitude bio_01 bio_02 bio_03 bio_04 bio_05 bio_06
#> 1        -97.75000        18.91667    155     92     57   2177    223     62
#> 2        -98.25000        18.75000    192    100     60   1990    264     97
#> 3        -98.58333        17.75000    210    104     62   1574    286    118
#> 4        -98.41667        19.41667    134     99     59   2204    205     38
#> 5        -96.58333        17.91667    204     80     53   2221    271    121
#> 6        -98.75000        18.41667    234    105     61   1801    311    139
#>   bio_07 bio_08 bio_09 bio_10 bio_11 bio_12 bio_13 bio_14 bio_15 bio_16 bio_17
#> 1    161    175    136    176    124    550    133      1      1    279      9
#> 2    167    210    177    211    163    690    187      0      1    408      7
#> 3    168    223    192    224    187    616    177      1      1    375      5
#> 4    167    156    103    156    103    613    119      0      1    326      8
#> 5    151    224    199    225    173   2520    619      4      1   1167     39
#> 6    172    248    214    250    208    611    158      0      1    380      3
#>   bio_18 bio_19
#> 1    259     14
#> 2    338      8
#> 3    296      5
#> 4    277      8
#> 5    972    583
#> 6    269      4The number of background points for each year is proportional to the number of occurrences for each year of observation.
Although the package uses minimum volume ellipsoids to model the niche, it has a function to export the time-specific data to Samples With Data format table that allows users to fit other algorithms such as MaxEnt. Let’s see how it works.
# SWD table for occurrence records
occ_swd <- tdf2swd(this_species=abex,sp_name="abro_gram")
# SWD table for background data
bg_swd <- tdf2swd(this_species=abbg)
head(tidyr::as_tibble(occ_swd))
#> # A tibble: 6 × 23
#>   sp_name   decimalLongitude decimalLatitude  year bio_01 bio_02 bio_03 bio_04
#>   <chr>                <dbl>           <dbl> <int>  <int>  <int>  <int>  <int>
#> 1 abro_gram            -97.3            18.7  1939    149     84     55   2252
#> 2 abro_gram            -97.3            18.7  1940    154     87     49   2575
#> 3 abro_gram            -97.0            19.6  1941    114     63     44   2693
#> 4 abro_gram            -97.3            18.7  1941    139     82     50   2394
#> 5 abro_gram            -97.3            18.7  1950    155     94     54   2079
#> 6 abro_gram            -97.1            19.7  1950    115     79     51   2278
#> # ℹ 15 more variables: bio_05 <int>, bio_06 <int>, bio_07 <int>, bio_08 <int>,
#> #   bio_09 <int>, bio_10 <int>, bio_11 <int>, bio_12 <int>, bio_13 <int>,
#> #   bio_14 <int>, bio_15 <int>, bio_16 <int>, bio_17 <int>, bio_18 <int>,
#> #   bio_19 <int>
head(tidyr::as_tibble(bg_swd))
#> # A tibble: 6 × 23
#>   sp_name    decimalLongitude decimalLatitude  year bio_01 bio_02 bio_03 bio_04
#>   <chr>                 <dbl>           <dbl> <dbl>  <int>  <int>  <int>  <int>
#> 1 background            -97.7            18.9  1939    155     92     57   2177
#> 2 background            -98.2            18.8  1939    192    100     60   1990
#> 3 background            -98.6            17.8  1939    210    104     62   1574
#> 4 background            -98.4            19.4  1939    134     99     59   2204
#> 5 background            -96.6            17.9  1939    204     80     53   2221
#> 6 background            -98.7            18.4  1939    234    105     61   1801
#> # ℹ 15 more variables: bio_05 <int>, bio_06 <int>, bio_07 <int>, bio_08 <int>,
#> #   bio_09 <int>, bio_10 <int>, bio_11 <int>, bio_12 <int>, bio_13 <int>,
#> #   bio_14 <int>, bio_15 <int>, bio_16 <int>, bio_17 <int>, bio_18 <int>,
#> #   bio_19 <int>As a final step, we will calibrate time-specific niche models using
minimum volume ellipsoids. To achieve this, we first select the
environmental variables using the function
tenm::correlation_finder. This function filters variables
according to a correlation threshold, which is important to avoid issues
related to collinearity.
varcorrs <- tenm::correlation_finder(environmental_data =
                                       abex$env_data[,-ncol(abex$env_data)],
                                     method = "spearman",
                                     threshold = 0.8,
                                     verbose = FALSE)
#> Warning in stats::cor(environmental_data, method = method): the standard
#> deviation is zero
# Selected variables
vars2fit <- varcorrs$descriptors
print(vars2fit)
#> [1] "bio_01" "bio_02" "bio_03" "bio_04" "bio_07" "bio_12" "bio_14" "bio_15"
#> [9] "bio_17"Now, we use the function tenm::tenm_selection to
calibrate the time-specific niche models. This function uses the
background object (here, the abbg object) as input. To
parametrize the function, we need to specify the omission rate criteria
to be used to select the models, the proportion of points to be included
in the ellipsoid model (ellipsoid_level parameter), the names
of the modeling layers (vars2fit parameter), a numeric vector
indicating the number of dimensions used to build ellipsoid models
(vars2fit parameter) a logical argument that determines whether
to use the partial ROC test or not, the random percent of data to be
used for the bootstrap of the partial ROC test (RandomPercent
parameter), the number of iterations of the partial ROC test
(NoOfIteration parameter), a logical argument to specify
whether to run the calibration process in parallel and the number of
cores used in the parallel process (parallel parameter).
mod_sel <- tenm::tenm_selection(this_species = abbg,
                                omr_criteria =0.1,
                                ellipsoid_level=0.975,
                                vars2fit = vars2fit,
                                nvars_to_fit=c(2,3,4,5,6,7),
                                proc = T,
                                RandomPercent = 50,
                                NoOfIteration=1000,
                                parallel=TRUE,
                                n_cores=4)
#> -----------------------------------------------------------------------------------------
#>      **** Starting model selection process ****
#> -----------------------------------------------------------------------------------------
#> 
#> A total number of 36 models will be created for combinations of 9 variables taken by 2 
#> 
#> A total number of 84 models will be created for combinations of 9 variables taken by 3 
#> 
#> A total number of 126 models will be created for combinations of 9 variables taken by 4 
#> 
#> A total number of 126 models will be created for combinations of 9 variables taken by 5 
#> 
#> A total number of 84 models will be created for combinations of 9 variables taken by 6 
#> 
#> A total number of 36 models will be created for combinations of 9 variables taken by 7 
#> 
#> -----------------------------------------------------------------------------------------
#>   **A total number of 492 models will be tested **
#> 
#> -----------------------------------------------------------------------------------------
#> Doing calibration from model  1 to  100 in process  1 
#> 
#> Doing calibration from model  101 to  200 in process  2 
#> 
#> Doing calibration from model  201 to  300 in process  3 
#> 
#> Doing calibration from model  301 to  400 in process  4 
#> 
#> Doing calibration from model  401 to  492 in process  5 
#> 
#> Finishing calibration of models  1 to  100 
#> 
#> Finishing calibration of models  101 to  200 
#> 
#> Finishing calibration of models  201 to  300 
#> 
#> Finishing calibration of models  301 to  400 
#> 
#> Finishing calibration of models  401 to  492 
#> 
#> Finishing...
#> 
#> -----------------------------------------------------------------------------------------
#>   244 models passed omr_criteria for train data
#>   27 models passed omr_criteria for test data
#>   27 models passed omr_criteria for train and test dataWe fitted 492 models, from which 27 passed our selection criteria.
Let’s explore the mod_sel object.
names(mod_sel)
#> [1] "temporal_df"  "sp_date_var"  "lon_lat_vars" "layers_ext"   "env_bg"      
#> [6] "mods_table"It has five slots. We can obtain the table of results of the
selection process by calling the mods_table slot.
head(mod_sel$mods_table,27)
#>                    fitted_vars nvars om_rate_train non_pred_train_ids
#> 1  bio_01,bio_02,bio_04,bio_07     4       0.06250              18,31
#> 2  bio_01,bio_02,bio_03,bio_04     4       0.06250              18,31
#> 3  bio_01,bio_03,bio_04,bio_07     4       0.06250              18,31
#> 4  bio_01,bio_04,bio_07,bio_12     4       0.09375           21,28,31
#> 5  bio_01,bio_02,bio_03,bio_07     4       0.03125                 18
#> 6         bio_01,bio_04,bio_07     3       0.06250              18,31
#> 7  bio_01,bio_03,bio_04,bio_12     4       0.09375           18,21,28
#> 8         bio_01,bio_03,bio_04     3       0.06250               3,18
#> 9                bio_01,bio_04     2       0.03125                 18
#> 10        bio_01,bio_02,bio_04     3       0.09375            3,18,31
#> 11               bio_01,bio_02     2       0.09375            3,18,31
#> 12               bio_01,bio_07     2       0.06250              18,31
#> 13               bio_01,bio_03     2       0.06250               3,18
#> 14        bio_01,bio_03,bio_12     3       0.06250               3,18
#> 15 bio_02,bio_04,bio_07,bio_12     4       0.06250              21,28
#> 16 bio_02,bio_03,bio_07,bio_12     4       0.06250              21,28
#> 17 bio_02,bio_03,bio_04,bio_12     4       0.06250              21,28
#> 18        bio_01,bio_07,bio_12     3       0.06250              18,31
#> 19 bio_02,bio_03,bio_04,bio_07     4       0.03125                  3
#> 20        bio_04,bio_07,bio_12     3       0.06250              21,28
#> 21               bio_04,bio_07     2       0.03125                  3
#> 22               bio_04,bio_12     2       0.06250              10,21
#> 23               bio_03,bio_04     2       0.03125                  3
#> 24        bio_02,bio_07,bio_12     3       0.09375            3,21,28
#> 25        bio_02,bio_03,bio_12     3       0.09375            3,21,28
#> 26        bio_03,bio_07,bio_12     3       0.09375            3,21,28
#> 27               bio_07,bio_12     2       0.03125                 28
#>    om_rate_test non_pred_test_ids bg_prevalence pval_bin pval_proc
#> 1             0                       0.4706024        0         0
#> 2             0                       0.4554527        0         0
#> 3             0                       0.4639894        0         0
#> 4             0                       0.4627871        0         0
#> 5             0                       0.4088013        0         0
#> 6             0                       0.4648311        0         0
#> 7             0                       0.4625466        0         0
#> 8             0                       0.4655525        0         0
#> 9             0                       0.4835878        0         0
#> 10            0                       0.4768546        0         0
#> 11            0                       0.4823855        0         0
#> 12            0                       0.5001804        0         0
#> 13            0                       0.4995792        0         0
#> 14            0                       0.4844295        0         0
#> 15            0                       0.6570879        0         0
#> 16            0                       0.5706385        0         0
#> 17            0                       0.6412168        0         0
#> 18            0                       0.4962126        0         0
#> 19            0                       0.6100757        0         0
#> 20            0                       0.6847421        0         0
#> 21            0                       0.7537574        0         0
#> 22            0                       0.7220151        0         0
#> 23            0                       0.7382470        0         0
#> 24            0                       0.6783696        0         0
#> 25            0                       0.6816160        0         0
#> 26            0                       0.7000120        0         0
#> 27            0                       0.7766021        0         0
#>    env_bg_paucratio env_bg_auc mean_omr_train_test rank_by_omr_train_test
#> 1          1.520879  0.7985825            0.031250                     11
#> 2          1.505433  0.7996963            0.031250                      7
#> 3          1.500235  0.7974813            0.031250                      8
#> 4          1.467522  0.7828762            0.046875                     22
#> 5          1.465208  0.7519787            0.015625                      1
#> 6          1.462961  0.7858313            0.031250                      9
#> 7          1.448599  0.7549675            0.046875                     21
#> 8          1.431383  0.7479400            0.031250                     10
#> 9          1.429927  0.7371300            0.015625                      2
#> 10         1.424194  0.7454525            0.046875                     23
#> 11         1.421552  0.7490962            0.046875                     24
#> 12         1.409348  0.7345675            0.031250                     15
#> 13         1.409145  0.7019625            0.031250                     14
#> 14         1.382147  0.6807450            0.031250                     12
#> 15         1.379969  0.7115000            0.031250                     18
#> 16         1.374939  0.6537675            0.031250                     16
#> 17         1.373252  0.7036712            0.031250                     17
#> 18         1.365017  0.7032725            0.031250                     13
#> 19         1.329613  0.6943437            0.015625                      3
#> 20         1.318842  0.6801412            0.031250                     19
#> 21         1.307749  0.6769513            0.015625                      5
#> 22         1.288655  0.6388300            0.031250                     20
#> 23         1.267623  0.6472100            0.015625                      4
#> 24         1.261588  0.6146950            0.046875                     25
#> 25         1.253496  0.5970800            0.046875                     26
#> 26         1.240056  0.5891425            0.046875                     27
#> 27         1.193064  0.5648687            0.015625                      6
#>    rank_omr_aucratio
#> 1                  1
#> 2                  2
#> 3                  3
#> 4                  4
#> 5                  5
#> 6                  6
#> 7                  7
#> 8                  8
#> 9                  9
#> 10                10
#> 11                11
#> 12                12
#> 13                13
#> 14                14
#> 15                15
#> 16                16
#> 17                17
#> 18                18
#> 19                19
#> 20                20
#> 21                21
#> 22                22
#> 23                23
#> 24                24
#> 25                25
#> 26                26
#> 27                27To project the models, we use the predict method. Here,
we will project one of the selected models using the environmental
layers of 2016. Also, we project it using layers from a period that
comprehends 1970-2000.
env_layers_2016 <- list.dirs(tempora_layers_dir,
                     recursive = FALSE)[32]
suit_2016 <- predict(mod_sel,
                     model_variables = c("bio_01","bio_03","bio_12"),
                     layers_path =env_layers_2016 ,
                     layers_ext = ".tif$")
#>   |                                                                              |                                                                      |   0%  |                                                                              |======================================================================| 100%Fig. 3. A selected niche model projected using environmental layers from 2016.
Now for the period that comprehends 1970-2000.
layers_70_00_dir <- system.file("extdata/bio_1970_2000",package = "tenm")
suit_1970_2000 <- predict(mod_sel,
                          model_variables = c("bio_01","bio_03","bio_12"),
                          layers_path = layers_70_00_dir,
                          layers_ext = ".tif$")
#>   |                                                                              |                                                                      |   0%  |                                                                              |======================================================================| 100%Fig. 4. A selected niche model projected using environmental layers from 1970-2000.
Lets see the predictions in geographic space
par(mfrow=c(1,2), mar=c(4,4,2,2))
terra::plot(suit_2016, main="Prediction for 2016")
terra::plot(suit_1970_2000, main="Prediction for 1970-2000")Fig. 5. Geographic projection of a selected model. Left panel, the projection using environmental layers from 2016. Right panel, the projection using environmental layers from 1970-2000
The following lines of code show the differences of a time-specific niche model and a standard niche model.
layers_70_00_dir <- system.file("extdata/bio_1970_2000",package = "tenm")
layers_70_00_path <- list.files(layers_70_00_dir,
                                pattern = ".tif$",full.names = TRUE)
# Extract environmental information 
elayers_70_00 <- terra::rast(layers_70_00_path)
e_trad <- terra::extract(elayers_70_00,
                         ab_1[,c("decimalLongitude","decimalLatitude")])
rgl::view3d(theta = 0, phi = -60,fov=120, zoom = 0.7) 
tenm::plot_ellipsoid(x = e_trad$bio_01,y=e_trad$bio_03,z=e_trad$bio_12,
                     col = "#1B9E77",
                     xlab = "Bio 1",
                     ylab = "Bio 3",
                     zlab = "Bio 12",)
tenm::plot_ellipsoid(x = abbg$temporal_df$bio_01,
                     y = abbg$temporal_df$bio_03,
                     z = abbg$temporal_df$bio_12,
                     col = "#E7298A",
                     add = TRUE)Fig. 6. Time-specific niche model vs. standard niche model. Pink ellipsoid represents the time-specific niche model. Green ellipsoid represents a ellipsoid model fitted using the standard approach.
Note that both ellipsoids differ in size and shape. In standard approach (green ellipsoid), we can see an sub-estimation of the environmental values where the intrinsic growth rate might be positive.