Background

I've slowly been creating Canadian-themed biodiversity projects over the past couple of years. Purely out of interest and to understand what my home and native land posts to iNaturalist.

The big project that is trying to be inclusive of all flora and fauna living on land and in water is found in my Biodiversity of Canada project.[1] My professional bias focuses my attention on ocean life, which is why I created my Marine Biodiversity of Canada umbrella project [2] and the individual ocean projects for each of Canada's three oceans.[3-5] With these projects, I could do what all good data-first folks do and started looking (playing? exploring?) and basic patterns of what the iNaturalist community observes (or not) in Canada.

Assumptions

It's fair to assume that most community-driven observations will be of organisms that live on land. The general public is mostly landlubbers, drylanders, and exploring on foot when casually being out and aboot. More specifically, water-proof technology is a luxury, niche item. Scuba-diving with a camera is an even more exclusive luxury skill. The fraction of ocean life that can be observed on shore is heavily influenced by the tides. The window of low tides throughout the year is pretty small which further restricts the slim chance most intertidal life will be observed. With all this in mind, the terrestrial observation bias was a given, but I still wanted to have some numbers to look at.

So, what are Canadians observing according to iNaturalist?

As of Mar. 8, 2024:

n= 13,012,012. The total biodiversity observations ([1])
n= 420, 696. The total marine biodiversity observations ([2])
n = 12,591,316. The total terrestrial biodiversity observations ([1] - [2])

97% of the biodiversity observations recorded on iNaturalist are of terrestrial flora and fauna. Or, a 30:1 ratio for how many terrestrial biodiversity observations are made to every marine biodiversity observation.

There's some grey to defining a species as either being terrestrial or marine. I was more liberal with including groups of species in the marine projects (e.g. waterfowl, some quadruped mammals, etc...). If excluded them from being classified as marine, then the terrestrial bias would be closer to 99% given more than 1/3 of the marine biodiversity observations are of seabirds!

It'd be interesting to dive a bit deeper into the taxonomic group bias and figure out which are the overlooked groups. I've already started exploring the one phylum that I've spent a couple of decades thinking about [6], but I have no doubt that organisms that have certain 'photographic traits' are going to be subconsciously selected for/against.

What does this mean?

These results probably shouldn't be surprising to traditional biodiversity scientists that operate independently of community science. The terrestrial > marine bias is often chatted about and not just in Canada. However, these numbers might be surprising to the iNaturalist community. If we consider that 1/3 of what we count as 'Canada' is ocean (land covers ~9.98 km² and ocean is ~5.75 km² ), the iNat community is definitely under-observing Canada's total real biodiversity pool with the largest potential observation gap being of what lives in and around salt water.

If I were to put myself in the shoes of somebody who just wants to (1) receive some Nature therapy, (2) improve the community data efforts on iNaturalist, (3) increase their chances of encountering a new (to them) animal, (4) and/or play the iNaturalist game of adding to their backyard species list...this first look at Canadian biodiversity data on iNat tells me that I could check off all those boxes in one afternoon by planning for a walk along the nearest beach or rocky intertidal habitat during the next decent low tide (<0.5 m during daylight).

[1] Biodiversity of Canada project
[2] Marine Biodiversity of Canada project
[3] Marine Biodiversity of Pacific Canada project
[4] Marine Biodiversity of Atantic Canada project
[5] Marine Biodiversity of Arctic Canada project
[6] Marine Sponges of the Pacific Northwest

*first posted in the project, Marine Sponges of the Pacific Northwest project within the umbrella Marine Biodiversity of Canada project.

Background:

The #1 assumption used by sponge biologists when identifying a sponge is that you can't reliably identify it with just a photo and without context. The scientific gold-standard for identifying a sponge is notoriously difficult to perform and involves taxonomic protocols that needs caustic chemicals, microscopy, and fluency in taxonomy jargon. Tell me the last time you heard somebody use "oxyhexaster microsclere" in a sentence that didn't involve sponge taxonomy?

As a biodiversity scientist and a sponge ecologist in a past life, I've continually thought about bridging old-school methods with new-school technology (e.g. iNaturalist). In other words, can we find some general rules that everybody can use to narrow down the identification process for what is paradoxically the most common marine animal in Earth's history but also the most ignored by most marine biologists?

I've dug through much of the iNat sponge observations in the Pacific Northwest over the past few years to get a sense of what are the most photographed sponge morphotypes (i.e., whole-sponge shape/color/form that may or may not represent a single species) where the science jury is still out on the taxonomy to allow these observations to be useful 'biodiversity data'. I feel a little bit of scientific detective work could go a long way with resolving the confidence in the ID of these common encounters in the rocky intertidal.

Ground rules (basic assumptions when identifying sponges):

(I) Sponges are the poster child of what it means to be 'phenotypically plastic'.

Many species are expected to have a diverse range of whole sponge morphotypes. As in, the same species could be a flat featureless, encrusting blob in some areas, but also be an erect, branching form in others. Very often, this is in response to the environment such as gradients of wave action between sites (low to high).[1]

(II) Diagnostic analyses of the presence, absence, and morphometrics of microscopic skeletal features (i.e., spicules) are still the gold-standard for identifying a sponge to species with the highest level of confidence.[2]

(III) Official species names for sponges are frequently being revised. In other words, their scientific names are constantly changing.

Some commonly photographed sponge morphotypes in the Pacific Northwest.

(I) The purple-encrusting, intertidal demosponge

Phylum: Porifera (high certainty)
Class: Demospongiae (high certainty)
Order: Haplosclerida (likely)

What the iNaturalist AI algorithm has commonly identified this as: Haliclona cinerea
What popular field guides have synonymized this as: Haliclona permollis, Haliclona cinerea, Reniera cinera, Reniera rufescens, Haliclona sp., Undetermined sponge.[3][4][5]

Sponge thoughts: Neither of the two most commonly applied species-level IDs used to identify this sponge on iNaturalist, Haliclona cinerea and H. permollis, are actually scientifically valid names for sponges. H. permollis is no longer recognized. H. cinerea is split with a subspecies to H. (Reniera) cinerea which appears to originate as a species described from waters around Britian.[6] It's rare to have scientific evidence that validates a marine species with having a "globally cosmopolitan distribution". The more common scenario I've encountered is that this had been an unvalidated assumption and upon further inspection, the extrapolated species is a 'new species'.

Regardless, the bigger issue is this morphotype is more than likely not a single species but a complex that includes several genus and species. The last I checked, we likely have undescribed species hiding within this complex on this coast - see this observations' comments for scientific dialogue/context (). It would be interesting to combine proper sponge taxonomy with an understanding of the established marine biogeographic/oceanography history of the Pacific Northwest for this morphotype. The true identities of this super common, yet understudied sponge morphotype. My hand-wavey prediction is that there is at least n=8 cryptic species with this morphotype occurring in the Pacific Northwest. I'm also speculating that 'purple' is not a fixed trait and that what can be purple can also be yellow, green, and beige.

(II) The red-encrusting, intertidal demosponge.

Phylum: Porifera (high certainty)
Class: Demospongiae (high certainty)
Order: Poecilosclerida (likely)
Family: Microcionidae (likely)
What the iNaturalist AI algorithm has commonly identified this as: Clathria pennata
What popular field guides have synonymized this as: Ophlitaspongia pennata, Desmacella pennata, Biemna pennata, Tylodesma pennata, Clathria pennata, Clathria sp., Undetermined sponge.[3][4]

Sponge thoughts: The genus Clathria is also complex with a diverse number of associated species.[7] I would like to guess that there are a number of cryptic species sharing this morphotype in the Pacific Northwest but this one seems to have gotten less attention over the years compared to the encrusting purple morphotype. The latter's plasticity and taxonomic chaos is much more well known.

Marine invertebrate zoology courses with field trips in the Pacific Northwest (e.g. British Columbia, Washington, Oregon) often go looking for animals at low tide. Instructors would often find this common bright red sponge as an example of the phylum Porifera and call it "Ophlitaspongia"while also mentioning the natural history note of the associated red dorid nudibranch Rostanga pulcra which is often camouflaged and feeding on this sponge. While the natural history fact is still accurate (see the photo for the red nudibranchs!), the genus Ophlitaspongia is not scientifically valid anymore. I'd be curious to know if R. pulcra feeds on all species with this morphotype or if it was selective?

Trailing thoughts

There are a few more sponge morphotypes that jump to mind (e.g. 'boot' sponges, Class hexactinellida) but I'll leave this post at two for now. One of my past research projects involved developing models that could predict areas of sponge habitat throughout the Pacific Northwest [8]. A task that required sourcing reliable sources of georeferenced, sponge observations (e.g. museum records). I briefly considered iNaturalist at the start but stopped with 'needs further work on the species IDs' when I encountered how inaccurate some of the common sponges were being done. Since then, there has been some improvements made by others - have a look at some of the efforts @tomleeturner to implement sponge identifying protocols using spicule analysis (and even DNA analysis) with iNaturalist photographs for each georeferenced individual observation as a template to emulate [9].

References:

[1] Palumbi, S.R., 1986. How body plans limit acclimation: responses of a demosponge to wave force. Ecology, 67(1), pp.208-214.

[2] Hooper, J.N. and Van Soest, R.W., 2002. Systema Porifera. A guide to the classification of sponges. In Systema Porifera: A guide to the classification of sponges (pp. 1-7). Boston, MA: Springer Us.

[3] Lamb, A. and Hanby, B.P., 2005. Marine life of the Pacific Northwest: a photographic encyclopedia of invertebrates, seaweeds and selected fishes. Harbour Publishing.

[4] Harbo, R.M., 2022. Whelks to whales: coastal marine life of the Pacific Northwest. Harbour Publishing.

[5] Invertebrates of the Salish Sea

[6] WoRMS taxon details - Haliclona (Reniera) cinerea (Grant, 1826)

[7] WoRMS taxon details - Clathria (Schmidt, 1862)

[8] Modelling the environmental niche space and distributions of cold-water corals and sponges in the Canadian northeast Pacific Ocean

[9] https://www.inaturalist.org/observations?place_id=any&project_id=marine-sponges-of-the-pacific-northwest&user_id=tomleeturner&verifiable=any

I decided to make a sub-project within my greater 'Marine Biodiversity of Canada" umbrella project to refocus some of my spare time on one group of animals that I've thought a lot about over the past couple of decades (https://www.inaturalist.org/projects/marine-sponges-of-the-pacific-northwest)

20 years ago, I stepped into my first sponge biology lab. My first lesson was that these animals are characterized by their 'absence' of everything. Sponges lack heads, organs, tissues. Some lack 'true cells'. They're often described as 'shapeless, amorphous blobs'. One encounter I had with professional oceanographers sticks with me - "I thought they were some kind of rock!". It made studying them hard when general ignorance was the predominant mindset among ocean science community. On the flipside, it was also rewarding when you can somehow generate some rules to push the science forward with baby steps.

For those open to learning, the first steps to identifying a sponge as a non-specialist is to accept that you can't identify them like you would a plant or a bird through macroscopic whole-organism features.

(1) You need microscopy to identify a sponge down to the species level with high confidence; the entire phylum's taxonomy is characterized by tiny skeleton features known as 'spicules'.

These glass or calcium carbonate 'lego bricks' make up the internal skeleton upon which the amorphous blob of tissues is draped. When the sponge dies (I also have stories to tell about how some oceanographers don't know how to tell the difference between a live, dead, and upside down sponge!), the spicules remain, but often the 'diagnostic spicules' are small, loose, and drift away with the decaying tissue. The presence or absence of these diagnostic spicules is how some species are identified. So, the false-absence of some spicules as a result of 'spicule loss' can also confuse even the best sponge taxonomists.

Therefore, the gold standard for a species-level ID for sponges requires a large fragment of the whole-healthy sponge for proper microscopy. A thorough inspection, with an exhaustive effort to look for 'rare and hard-to-find diagnostic spicules', also makes high-level taxonomic work difficult. Without microscopic evidence and focused detective-type lab work, any species-level sponge ID would be an informed best guess by even professionally trained sponge biologists (i.e. take it with a grain of salt, and with a large degree of uncertainty).

(2) Sponges are known to be diverse in form, meaning a species of sponge can take the whole-body shape of a saucer, a boot, or a blob.

However,

(3) Sponges are known for their cryptic diversity, meaning several species of sponges can look identical.

Endless arguments from sponge taxonomists with sponge 'biologists' have centered on whether or not we can move away from high rigor and have 'looser' IDs (heated debates).

Together, this adds to the difficulty of bridging the heart of how iNaturalist crowd-sources 'species identifications' with easy-to-capture photographs of the organism to generate hypothetical data that could be useful for scientists like me who research/map biodiversity patterns.

At the moment, I won't use iNaturalist data for this group of animals because the crowd-sourcing process forces 'incorrect' species IDs but labels it as 'research-grade'. In other words, the iNat platform has systematically injected wrong IDs into the majority of the data because the majority crowd-sourced audience doesn't have formal official knowledge to correctly ID an entire phylum's worth of species.

This is by no means saying iNaturalist isn't useful (I have the complete opposite perspective), but rather an example of how 'marine' biodiversity observations on iNat are lagging behind terrestrial efforts (deja vu if you've followed the history of ecological research).

So, using sponges as an example, are there rules that can help intertidalists and scuba divers put a name to their wonderful photos of sponges on iNaturalist?

(i) The auto-suggestion algorithms for iNaturalist for sponges are more than likely incorrect. However, with systematic incorrect labelling, projects like this can hopefully back-trace a more appropriate species ID in the future.

(ii) When taking snapshots, take a picture of the whole sponge and provide as accurate a location as possible. At a minimum, one can assume the same morphotype from a 'geographically distinct' population will be the same species, so a voucher specimen sampled at a later date might improve ID confidence for a commonly visited site.

(iii) There are three 'common' classes of sponges: Demosponges, Glass Sponges, and Calcarea Sponges (stop right there with Homosclereomorphs - i said common and this post is about the Pacific Northwest).

Without expert training, it's more than likely any intertidal sponge will either be a demosponge or a calcareous sponge. Glass sponges are only found in deep waters at the edge of safe recreational scuba diving limits. Calcareous sponges and glass sponges also have much less morphological variety when compared to demosponges.

So, through a process of elimination, if it doesn't look like the standard calcareous sponge (small finger shaped or bird's nest intertwined forms) or glass sponge (cloud sponges bushes and globlets, boot sponge cylinders), it'll likely be a demosponge (every infinite permutation of amorphous blob you can think of). Therefore, most IDs can be done with high confidence to the class level by everyday explorers.

Hopefully, this helps improve the data being generated by community scientists for this group of animals. Sponges have survived every mass extinction our planet has experienced (and will survive the Anthropocene). They're one of the oldest animals found in the fossil record. They have occupied every aquatic habitat on our planet They are found in every ocean from pole to pole. Our collective science has historically misrepresented this group as a 'minority' of the marine biodiversity world, but in a reality, they are the dominant animal of our past, present, and future planet.

I'm always super appreciative of the expansive knowledge and a curiously that comes from the iNaturalist community. So, I wanted to start keeping track of the fun and very informative conversations that resulted from some of my more unusual observations.

I'm slowly pulling most of the ones that I can remember, but I'll be adding more as discussions continue to pop up over time.

• Deep-sea coral or parasite egg?
• Your purple sponge is likely not Haliclona cinerea?
• Sometimes a photo just isn't enough for a species-level ID?
• Don't forget the headshot?
• What deep-sea creature is in this video?
• How to deduce an ID with no previous photos on iNaturalist?
• Didn't think sablefish were found inshore?
• Deep-sea bacteria - you know you want to know more?
• I've seen those corals before! from 50 years ago?