In monteroserra/rOceans: Spatial Trends in Marine Biodiversity and Conservation

rOceans (GSoC2018) by Ignasi Montero-Serra

This file describes the recently developed rOceans package, the datapackage rOceansData and contains a study case on marine benthic biodiversity. The project has been lead by Ignasi Montero-Serra and fundend by Google within the program Google Summer Code 2018.

Deliverable 1 rOceans

rOceans is an R Package for exploring spatial trends in marine biodiversity and conservation. It contains functions to facilitate the accessibility and exploration of spatial trends of marine biodiversity, current environmental drivers and future global change scenarios.

Deliverable 2 rOceansData

rOceansData is an R DataPackage that contains several marine biodiversity layers and environmental drivers to make easy explorations using rOceans. It is stored in github.com/monteroserra/rOceansData

Deliverable 3 rOceansApp

Is an R Shiny App that alllos an easy visualization of some examples of marine biodiversity computed using rOceans. The App can be explored at:

https://monteroserra.shinyapps.io/rOceansApp/

rOceans

Functions list

# Functions' name Description

---

(1) oceanDataAccess access occurrences data from OBIS and GBIF for a list of species and provides options for data quality checks (2) oceanDataCheck automatically merges, checks and filters large datasets directly downoladed from OBIS and GBIF (3) oceanAbundance computes global or regional spatial grids of abundance of occurrences per grid cell at multiple scales (4) oceanMaps versatile visualization tool for representing spatial trends in biodiversity (5) oceanDiversity computes spatial explicit layers of several diversity metrics including richness, corrected richness, Simpson and Shannon diversity (6) oceanEnvironment access to data on global spatial layers of multiple environmental drivers and explores relationships to marine biodiversity metrics (7) oceanFuture access current and future environmental data layers and computes expected trends (8) oceanHotspots classify sites in hotspots and coldspots of biodiversity (9) oceanVulnerability explores and ranks vulnerability of sites by linking biodiversity classifications and expected warming trends

---

Download rOceans from GitHub and load the package

#devtools::install_github("monteroserra/rOceans")
#devtools::install_github("monteroserra/rOceansData")
library(rOceans)
library(rOceansData)

Function 2 oceanDataCheck

This functions allows merging, checking, filtering raw data directly downloaded from OBIS and/or GBIF and creates a common standard database. It removes occurrences with inconsistent or inconmplete taxonomic information, non-marine (land-based) occurrences, and duplicate occurrences.

For this set of functions, we first need to access and download data from GBIF (www.gbif.org) and OBIS (www.iobis.org).

Here I show an example with data of species within the genus Acropora that were directly downloaded from the website on June 2018. The data can be accessed from my GitHub repository "monteroserra/rOceanData" or using data()

#Accesing raw datasets
data(Acropora_GBIF) 
data(Acropora_OBIS)

#Checking OBIS or GBIF individually
Acropora_OBIS_checked = oceanDataCheck(OBIS_occurrences = Acropora_OBIS, source = "OBIS")

Acropora_GBIF_checked = oceanDataCheck(GBIF_occurrences = Acropora_GBIF, source = "GBIF")

This function will tell us how many occurrences were removed due to issues with the coordinates (containning NAs), land-based occurrences, and duplicate occurrences between OBIS and GBIF davases.

Merging OBIS and GBIF Datasets, and checking for taxonomic and geographic issues

Acropora_Total_Checked = oceanDataCheck(GBIF_occurrences = Acropora_GBIF,
                                        OBIS_occurrences = Acropora_OBIS,
                                        source = "GBIF_&_OBIS")

Function 3 oceanAbundance

Allows creating global or delimited spatial grids with abundance of occurrences per cell at differenc cell sizes.

Following with the Acropora example, we will use this function to compute global rasters of abundance per grid cell for Acropora species at different resolutions (10 x 10) (5 x 5) (1 x 1) and (0.5 x 0.5) using cell_size

Acropora_abundance_10x10 = oceanAbundance(occurrences = Acropora_Total_Checked, cell_size=10)
Acropora_abundance_5x5 = oceanAbundance(occurrences = Acropora_Total_Checked, cell_size=5)
Acropora_abundance_1x1 = oceanAbundance(occurrences = Acropora_Total_Checked, cell_size=1)
Acropora_abundance_0.5x0.5 = oceanAbundance(occurrences = Acropora_Total_Checked, cell_size=0.5)

Function 4 oceanMaps

A versatile data mapping tool for a rapid visualization of marine biodiversity patterns previously computed.

par(mfrow=c(2,2))
oceanMaps(Acropora_abundance_10x10, logScale=T, main="Acropora abundance (10x10)")
oceanMaps(Acropora_abundance_5x5, logScale=T, main="Acropora abundance (5x5)")
oceanMaps(Acropora_abundance_1x1, logScale=T, main="Acropora abundance (1x1)")
oceanMaps(Acropora_abundance_0.5x0.5, logScale=T, main="Acropora abundance (0.5x0.5)")
par(mfrow=c(1,1))

Study Case: global patterns in benthic biodiversity

We will explore the abundance and diversity of eight major benthic groups with data downloaded from GBIF and OBIS. The data stored in rOceansData has already preprocessed by merging both datasets (GBIF and OBIS) and filtering and removing land-based occurrences and data with incomplete taxonomic infor using the function oceansDataCheck() The main groups included in this study case are:

Anthozoans, Ascidians,Bryozoans, Porifers, Bivalves, Macroalgae, Seagrasses and sedentary Polychaets.

Load the data from rOceansData

library(rOceansData)

data("Anthozoa_Total_Checked")
data("Bryozoa_Total_Checked")
data("Porifera_Total_Checked")
data("Ascidiacea_Total_Checked")
data("Bivalvia_Total_Checked")
data("Cirripeda_Polychaeta_Total_Checked")
data("Macroalgae_Total_Checked")
data("Seagrass_Total_Checked")

Now we compute abundance trends

Anthozoa_abundance = oceanAbundance(occurrences = Anthozoa_Total_Checked)
Bryozoa_abundance = oceanAbundance(occurrences = Bryozoa_Total_Checked)
Porifera_abundance = oceanAbundance(occurrences = Porifera_Total_Checked)
Ascidiacea_abundance = oceanAbundance(occurrences = Ascidiacea_Total_Checked)
Bivalvia_abundance = oceanAbundance(occurrences = Bivalvia_Total_Checked)
Cirripeda_Polychaeta_abundance = oceanAbundance(occurrences = Cirripeda_Polychaeta_Total_Checked)
Macroalgae_abundance = oceanAbundance(occurrences = Macroalgae_Total_Checked)
Seagrasses_abundance = oceanAbundance(occurrences = Seagrasses_Total_Checked)

We can visualize the total occurrence of benthic species

Marine_sessile_total = rbind(
              Anthozoa_Total_Checked,
              Bryozoa_Total_Checked,
              Porifera_Total_Checked,
              Ascidiacea_Total_Checked, 
              Bivalvia_Total_Checked,
              Cirripeda_Polychaeta_Total_Checked,
              Seagrasses_Total_Checked,
              Macroalgae_Total_Checked)


Marine_sessile_abundance = oceanAbundance(occurrences =Marine_sessile_total)

oceanMaps(Marine_sessile_abundance, logScale=T, background_color = "grey70", main = "All species abundance", low_color = "#B4EEB4", mid_color = "#6959CD", high_color = "#9B3F2F")

This global layer can be used as a proxy for sampling effort in any in-depth analysis of benthic biodiversity using data from OBIS and GBIF.

Visualizing abundance for each group

par(mfrow=c(2,2))

oceanMaps(Anthozoa_abundance, logScale=T, background_color = "grey70", main = "Anthozoa",
          low_color = "#FFF8DC", mid_color = "#CD5B45", high_color = "#9B3F2F")

oceanMaps(Bryozoa_abundance, logScale=T, background_color = "grey70", main = "Bryozoa",
          low_color = "#FFF8DC", mid_color = "#CD5B45", high_color = "#9B3F2F")

oceanMaps(Ascidiacea_abundance, logScale=T, background_color = "grey70", main = "Ascidiacea",
          low_color = "#B4EEB4", mid_color = "#6959CD", high_color = "#473C8B")

oceanMaps(Porifera_abundance, logScale=T, background_color = "grey70", main = "Porifera",
          low_color = "#B4EEB4", mid_color = "#6959CD", high_color = "#473C8B")

oceanMaps(Bivalvia_abundance, logScale=T, background_color = "grey70", main = "Bivalvia",
          low_color = "#FFF8DC", mid_color = "#CDB79E", high_color = "#8D7B6D")

oceanMaps(Cirripeda_Polychaeta_abundance, logScale=T, background_color = "grey70", main = "Cirripeda & Polychaeta",
          low_color = "#FFF8DC", mid_color = "#CDB79E", high_color = "#8D7B6D")

oceanMaps(Macroalgae_abundance, logScale=T, background_color = "grey70", main = "Macroalgae",
          low_color = "#CAFF70", mid_color = "#A2CD5A", high_color = "#006400")

oceanMaps(Seagrasses_abundance, logScale=T,background_color = "grey70", main = "Seagrasses",
          low_color = "#CAFF70", mid_color = "#A2CD5A", high_color = "#006400")


par(mfrow=c(1,1))

Exploring diversity patterns

Diversity can be computed using the function oceanDiversity() for instance:

#Anthozoa_diversity = oceanDiversity(occurrences = Anthozoa_Total_Checked)

For simplicity, here we present a set of preprocessed diversity objecst of the major benthic groups, which are stored in the rOceansData package:

Here we can observe the major patterns in species richness

data("Anthozoa_diversity")
data("Bryozoa_diversity")
data("Ascidiacea_diversity")
data("Porifera_diversity")
data("Bivalvia_diversity")
data("Cirripeda_Polychaeta_diversity")
data("Macroalgae_diversity")
data("Seagrass_diversity")

par(mfrow=c(2,2))

oceanMaps(Anthozoa_diversity[[2]], logScale=T, background_color = "grey70", main = "Anthozoa",
          low_color = "#FFF8DC", mid_color = "#CD5B45", high_color = "#9B3F2F")

oceanMaps(Bryozoa_diversity[[2]], logScale=T, background_color = "grey70", main = "Bryozoa",
          low_color = "#FFF8DC", mid_color = "#CD5B45", high_color = "#9B3F2F")

oceanMaps(Ascidiacea_diversity[[2]], logScale=T, background_color = "grey70", main = "Ascidiacea",
          low_color = "#B4EEB4", mid_color = "#6959CD", high_color = "#473C8B")

oceanMaps(Porifera_diversity[[2]], logScale=T, background_color = "grey70", main = "Porifera",
          low_color = "#B4EEB4", mid_color = "#6959CD", high_color = "#473C8B")

oceanMaps(Bivalvia_diversity[[2]], logScale=T, background_color = "grey70", main = "Bivalvia",
          low_color = "#FFF8DC", mid_color = "#CDB79E", high_color = "#8D7B6D")


oceanMaps(Cirripeda_Polychaeta_diversity[[2]], logScale=T, background_color = "grey70", main = "Cirripeda & Polychaeta",
          low_color = "#FFF8DC", mid_color = "#CDB79E", high_color = "#8D7B6D")

oceanMaps(Macroalgae_diversity[[2]], logScale=T, background_color = "grey70", main = "Macroalgae",
          low_color = "#CAFF70", mid_color = "#A2CD5A", high_color = "#006400")

oceanMaps(Seagrasses_diversity[[2]], logScale=T,background_color = "grey70", main = "Seagrasses",
          low_color = "#CAFF70", mid_color = "#A2CD5A", high_color = "#006400")

And species richness applying a correction tools called rarefaction with a treshold sample size of 50 samples

par(mfrow=c(2,2))

oceanMaps(Anthozoa_diversity[[3]], logScale=T, background_color = "grey70", main = "Anthozoa",
          low_color = "#FFF8DC", mid_color = "#CD5B45", high_color = "#9B3F2F")

oceanMaps(Bryozoa_diversity[[3]], logScale=T, background_color = "grey70", main = "Bryozoa",
          low_color = "#FFF8DC", mid_color = "#CD5B45", high_color = "#9B3F2F")

oceanMaps(Ascidiacea_diversity[[3]], logScale=T, background_color = "grey70", main = "Ascidiacea",
          low_color = "#B4EEB4", mid_color = "#6959CD", high_color = "#473C8B")

oceanMaps(Porifera_diversity[[3]], logScale=T, background_color = "grey70", main = "Porifera",
          low_color = "#B4EEB4", mid_color = "#6959CD", high_color = "#473C8B")

oceanMaps(Bivalvia_diversity[[3]], logScale=T, background_color = "grey70", main = "Bivalvia",
          low_color = "#FFF8DC", mid_color = "#CDB79E", high_color = "#8D7B6D")

oceanMaps(Cirripeda_Polychaeta_diversity[[3]], logScale=T, background_color = "grey70", main = "Cirripeda & Polychaeta",
          low_color = "#FFF8DC", mid_color = "#CDB79E", high_color = "#8D7B6D")

oceanMaps(Macroalgae_diversity[[3]], logScale=T, background_color = "grey70", main = "Macroalgae",
          low_color = "#CAFF70", mid_color = "#A2CD5A", high_color = "#006400")

oceanMaps(Seagrasses_diversity[[3]], logScale=T,background_color = "grey70", main = "Seagrasses",
          low_color = "#CAFF70", mid_color = "#A2CD5A", high_color = "#006400")

Getting current layers of envioronmental parameters and linking them to biodiversity patterns

oceanEnvironment(biod_grid = Anthozoa_diversity[[3]],
                             biodiv_metric = "Correced Richness", 
                             plot=T, 
                             log_scale=T)

Now we can use te function oceanHotspots() to classify biodiversity sites in high (red); mid(yellow) and low (blue) biodiversity

par(mfrow=c(2,2))

oceanHotspots(Anthozoa_diversity[[3]], map_hotspots = T)
oceanHotspots(Bryozoa_diversity[[3]], map_hotspots = T)
oceanHotspots(Macroalgae_diversity[[3]], map_hotspots = T)
oceanHotspots(Seagrasses_diversity[[3]], map_hotspots = T)

Finally, the function oceanVulnerability() allows us to use future climate change scenarios, that can be accessed using oceanFuture() and explore potential climatic vulenrability of high biodiversity sites.

par(mfrow=c(2,2))


oceanVulnerability(Anthozoa_diversity[[3]], plot_histograms = F)

oceanVulnerability(Seagrasses_diversity[[3]], plot_histograms = F)

These maps show potential vulnerability rankings of hotspots of biodiversity in Anthozoans and Seagrasses

monteroserra/rOceans documentation built on Dec. 23, 2024, 2:47 p.m.