flowchart LR
%% nodes
P(presence):::in
A("pseudo-absence"):::in
pts(obs):::out
X[/extract/]:::fxn
env(env):::in
D[[data]]:::out
S[/split/]:::fxn
t1[[train]]:::out
t2[[test]]:::out
F[/fit/]:::fxn
M{{model}}:::out
E[/evaluate/]:::fxn
C[/calibrate/]:::fxn
pr[/predict/]:::fxn
N[["new data"]]:::in
prd(prediction):::out
prf(performance):::out
res[result set]:::out
ens[ensemble]:::fxn
der[derived]:::out
web[web]:::out
%% edges
P --> pts
A --> pts
pts --> X
env --> X
X --> D
D --> S
S --> t1
S --> t2
t1 --> F
F --> M
t2 --> E
M --> C
C --> F
M --> pr
N --> pr
M --> E
pr --> prd
E --> prf
prd --> res
prf --> res
res --x ens
ens --> der
der --> web
res --> web
%% subgraphs
subgraph Share
direction LR
subgraph Combine
subgraph Prepare
P
A
pts
env
X
D
end
subgraph Model
S
t1
t2
F
M
E
C
pr
N
prd
prf
res
end
ens
der
end
web
end
%% Legend
in(input):::in
fxn[/process/]:::fxn
out(output):::out
subgraph Legend
in
fxn
out
end
%% styles
classDef in fill:lightblue
classDef out fill:lightgreen
classDef fxn fill:yellow
MarineSDMs
Marine Species Distribution Models
1 Introduction
1.1 Background
The best available global distributions are presently AquaMaps (Kaschner et al. 2006; Ready et al. 2010) with supplementation by IUCN RedList range maps1. These have been used to calculate the biodiversity within national waters (Halpern et al. 2012) as well as beyond in the high seas (Visalli et al. 2020).
1.2 Goals
This book aims to capture the overview and details of modeling species distributions in the marine environment for the purposes of advancing the status quo of global and U.S. national species distributions along the following dimensions:
Space
The current AquaMaps distributions are \({1}/{2}\)º (~55 km at equator), whereas the best available global bathymetry is \({1}/{240}\)º (< 0.5 km).Time
The current AquaMaps distributions are based on static climatic averages over all seasons, which does not capture temporal dynamics: seasonally within a year, nor long-term climate change trends. This will necessitate sampling the environment contemporaneously with species observations before fitting the model and predicting to different environmental snapshots.Environment
Other environmental variables besides the initial physiographic (depth) and oceanographic (temperature, chlorophyll, primary productivity and ice) may elicit an improved statistical fit, related to species’ environmental niche. Some candidates include: temperature fronts, eddy kinetic energy, distance from shore, distance from shelf.Biology
Where sufficient observations exist, additional models should be developed highlighting differences between:Life stage, e.g. larval vs adult.
Gender where varies, such as male sperm whales being more cosmopolitan.
Subpopulations for understanding metapopulation dynamics
Behavior, such as migrating, feeding or breeding.
By definition MBONMarine Biodiversity Observation Network; see MarineBON.org is a network, so this is inclusive of and meant for all participants.
1.3 Motivations
AquaMaps.org
AquaMaps (Kaschner et al. 2006; Ready et al. 2010) represents a massive amount of work to gather parameters for >33.5K marine species, including areas to mask out.OBIS.org
The Ocean Biogeographic Information System (Klein et al. 2019; Grassle 2000) is the central portal for continuously added observations with extra flags for quality control, all of which makes marine SDMs possible.Modeling methods have dramatically improved over time and are ripe for fresh application. The R package
dismooriginally came came out with an SDM vignette as a practical supplement to their excellent review of SDMs (Elith and Leathwick 2009) and using the Maxent algorithm (Elith et al. 2011). The raster package furthered that (raster sdm) and now there’sterrasdm. Alongside these developments has been a boon of cloud-computing, particularly Google Earth Engine (Gorelick et al. 2017; Campos et al. 2023), allowing for dense global raster processing.The world is quickly moving towards a future trying to conserve 30% of the oceans by 2030, so called “30 by 30”. In the U.S., this is America the Beautiful initiative (Carroll, Noss, and Stein 2022) for which MBONMarine Biodiversity Observation Network; see MarineBON.org is well poised to inform (Fautin et al. 2010; Muller-Karger et al. 2018). We need biodiversity indicators to track progress. This push for conservation is driven by increasing impacts of climate change, as evidenced by marine heatwaves and shifts in population distributions.
1.4 Process
1.5 Contribute
We very much welcome your feedback, contributions and collaboration. As soon as you contribute, we will add you to to the authors list. Here are a few ways to contribute from least to most involved:
Email Ben (ben@ecoquants.com) with any suggestions, including suggested revisions of this online book.
Submit a New Issue on Github.
Click on “ Edit this Page” in the upper right. If you have a Github account, then you can fork this repository from owner “marinebon” to your username, edit the page(s) and submit a pull request. See Hello World - GitHub Docs.
If you are a regular contributor, you can be added to the collaborators of this repository to push changes directly (without needing a pull request).
1.6 Features
This Quarto book has a few cool features:
Multiple formats
From the singe set of source Quarto documents (*.qmd), several output formats are rendered: html, pdf, docx. This is particularly helpful when suggesting changes. It also lends itself well to being carved into manuscripts.Self-rendering
Github hosts the web pages (*.html), which get rendered from the source code (*.qmd) using a Github Action. So edits can be made simply through the web interface and all outputs get updated (html, pdf, docx). It also ensures the reproducibility of the document with a common setup environment.Mermaid diagrams
e.g., Figure 1.1, Figure 3.1, Figure 6.1Quarto document listings
References
Glossary
Search
IUCN RedList range maps: https://www.iucnredlist.org/resources/spatial-data-download↩︎