Case study 1: Different biodiversity hotspots and their congruence.

Project Set Up

Initiates a new rangeMapper project (wrens.sqlite) into a temporary directory

td = tempdir()
require(rangeMapper)
dbcon = rangeMap.start(file = "wrens.sqlite", dir = td, overwrite = TRUE)

## New session 2016-01-01 17:39:52
## PROJECT: wrens.sqlite 
## DIRECTORY: /tmp/RtmpQXAmRj

# Location of the vector(*.shp) breeding ranges on disk
branges = system.file(package = "rangeMapper", "extdata", "wrens", "vector_combined")

Saves the global bounding box as the union of all species bounding boxes.

global.bbox.save(con = dbcon, bbox = branges)

## Computing global bounding box for 1 ranges...
## Checking for proj4 string differences...
## Done!
## Bounding box uploaded. PROJ4STRING set to  +proj=longlat +datum=NAD83 +no_defs

Saves the grid size (i.e. the size of a canvas cell) using the default value. The grid size can be specified by e.g. gridSize.save(dbcon, gridSize = 2.5)

gridSize.save(dbcon)

## Warning in .local(object, ...): Default grid size used!

## Grid size set to 1.17283756256104 map units.

Saves the canvas grid using the global bounding box and the grid size.

canvas.save(dbcon)

## Canvas uploaded.

Performs vector range maps interpolation with the canvas.

r = rgdal::readOGR(branges, "wrens", verbose = FALSE)
processRanges(spdf = r, con =  dbcon, ID = "sci_name", metadata = rangeTraits() )

## Warning in .local(con, spdf, ID, metadata): Reprojecting to "+proj=longlat
## +datum=NAD83 +no_defs "

## Writing to project.
## 84 ranges updated; Elapsed time: 0.9 secs
## Extracting metadata...

## [1] TRUE

Uploads to the existing project a data.frame containing life history data

data(wrens)
bio.save(con = dbcon, loc = wrens, ID = "sci_name")

## Table wrens saved as a  BIO_ table

Define Hotspots Parameters

(i.e. the quantile of the given parameter used as a threshold value)

P_richness = 0.75 # species richness probability
P_endemics = 0.35 # endemic species richness probability

# 1) Save the species richeness (SR) map
rangeMap.save(dbcon, tableName = "species_richness")

## [1] TRUE

# 2) Fetch the SR map and find the SR value corresponding with the probability
 # previously defined by P_richness
sr = rangeMap.fetch(dbcon, "species_richness")
sr_threshold = quantile(sr$species_richness, probs = P_richness, na.rm = TRUE)

# 3) Save the SR hotspots map; only canvas cells with species_richness >= sr_threshold will be included
rangeMap.save(dbcon, tableName = "species_richness_hotspots",
    subset = list(MAP_species_richness =
        paste("species_richness >=", sr_threshold)) )

## [1] TRUE

# 1) Save endemics species richnes map
es = RSQLite::dbGetQuery(dbcon, "select Area from metadata_ranges")
es_threshold = quantile(es$Area, probs = P_endemics, na.rm = TRUE)

rangeMap.save(dbcon, tableName = "endemics_spRichness",subset =
    list(metadata_ranges = paste("Area <= ", es_threshold) ) )

## [1] TRUE

# 2) Save hotspots of endemics
er = rangeMap.fetch(dbcon, "endemics_spRichness")
er_threshold = quantile(er$endemics_spRichness, probs = P_richness, na.rm = TRUE)

rangeMap.save(dbcon, tableName = "endemics_hotspots", subset = list(
    MAP_endemics_spRichness = paste("endemics_spRichness >=", er_threshold) ) )

## [1] TRUE

#1) Save Coefficient of variation (CV_Mass) map using a function (FUN) defined
 # on the fly. The ... argunent of the function will allow further arguments
 # in this case 'na.rm = TRUE' to be passed to FUN.

rangeMap.save(dbcon,
    FUN = function(x,...) (sd(log(x),...)/mean(log(x),...)),
    na.rm = TRUE, biotab = "wrens", biotrait = "body_mass",
    tableName = "CV_Mass")

## [1] TRUE

# 2) Find the 'CV_Mass' threshold which corresponds with 'P_richness'
bmr = rangeMap.fetch(dbcon, "CV_Mass")
bmr_threshold = quantile(bmr$CV_Mass, probs = P_richness, na.rm = TRUE)

# 3) Save bodyMass richness hotspots map using the computed threshold
rangeMap.save(dbcon, tableName = "bodyMass_richness_hotspots",
    subset = list(MAP_CV_Mass =
        paste("body_mass >=", bmr_threshold) ) )

## [1] TRUE

#  FETCH MAPS & PLOT
SRmaps = rangeMap.fetch(dbcon, c("species_richness", "species_richness_hotspots",
    "endemics_spRichness", "CV_Mass"), spatial = FALSE )
plot(SRmaps)

HOTSPOTS CONGRUENCE

1. Find threshold values

sr = rangeMap.fetch(dbcon, "species_richness")
Q_richnes = quantile(sr$species_richness, probs = P_richness, na.rm = TRUE)

er = rangeMap.fetch(dbcon, "endemics_spRichness")
Q_endemics = quantile(er$endemics_spRichness, probs = P_richness, na.rm = TRUE)

bmr = rangeMap.fetch(dbcon, "CV_Mass")
Q_bodyMass = quantile(bmr$CV_Mass, probs = P_richness,na.rm = TRUE)

2. Save cumulative hotspot congruence satisfying all three criteria

rangeMap.save(dbcon, tableName = "cumul_congruence_hotspots",subset = list(
    MAP_CV_Mass             = paste("body_mass>=",Q_bodyMass),
    MAP_species_richness    = paste("species_richness>=",Q_richnes),
    MAP_endemics_spRichness = paste("endemics_spRichness>=",Q_endemics) ))

## [1] TRUE

3. Species richness & endemics richness congruence (two criteria are satified)

rangeMap.save(dbcon, tableName = "SR_ER_congruence_hotspots",subset = list(
    MAP_species_richness    = paste("species_richness>=",Q_richnes),
    MAP_endemics_spRichness = paste("endemics_spRichness>=",Q_endemics) ))

## [1] TRUE

4. Endemics richness & body mass diversity congruence

rangeMap.save(dbcon, tableName = "ER_BR_congruence_hotspots",
     subset = list(
     MAP_CV_Mass=paste("body_mass>=",Q_bodyMass),
     MAP_endemics_spRichness=paste("endemics_spRichness>=",Q_endemics) ))

## [1] TRUE

5. Species richness & body mass diversity congruence

rangeMap.save(dbcon, tableName = "SR_BR_congruence_hotspots",
     subset = list(
     MAP_CV_Mass=paste("body_mass>=",Q_bodyMass),
     MAP_species_richness=paste("species_richness>=",Q_richnes) ))

## [1] TRUE

6. Retrieve all the map objects.

m = rangeMap.fetch(dbcon, c("species_richness_hotspots","endemics_hotspots",
    "bodyMass_richness_hotspots","cumul_congruence_hotspots",
    "SR_ER_congruence_hotspots","SR_BR_congruence_hotspots",
    "ER_BR_congruence_hotspots") )

7. Compute hotspots congruence

# The data slot of the map 'm' can be queried for presence/absence using the 'is.na' function.

sr = length( which(!is.na(m$species_richness_hotspots)))
er = length( which(!is.na(m$endemics_hotspots)))
br = length( which(!is.na(m$bodyMass_richness_hotspots)))

srer = length( which(!is.na(na.omit(m$SR_ER_congruence_hotspots))) )
srbr = length( which(!is.na(na.omit(m$SR_BR_congruence_hotspots))) )
erbr = length( which(!is.na(na.omit(m$ER_BR_congruence_hotspots))) )

cum = length( which(!is.na(na.omit(m$cumul_congruence_hotspots))) )

Species richness & body mass diversity congruence (%)

(srbr/(sr+br))*100

## [1] 24.34861

Endemics richness & body mass diversity congruence (%)

(erbr/(er+br))*100

## [1] 4.779412

Species richness & endemics richness congruence (%)

(srer/(sr+er))*100

## [1] 6.695779

Cumulative congruence (%)

(cum/(sr+er+br))*100

## [1] 1.962457