This week iNaturalist passed another big observation milestone: 200,000,000 verifiable observations! It took iNaturalist 14 years to reach 100 million and just 2 more years to reach 200 million.

If each observation were a coin, the stack would be almost 200 miles high.

If each observation were a minute, we could spin the clock back to the year 1644.

If each observation were a horse, that would be 3 times as many horses as are on Earth today.

Let’s take a closer look at these observations as 200 dots where each represents 1 million observations.

Imagine 200 million observations are represented by 200 dots

133 of the 200 dots are of plants and insects. Fish are represented by just 2 dots.

How many dots from each species category?

114 of the 200 dots are from in North America. Africa and South America have the fewest dots.

How many dots are from each continent?

The graph below shows the number of observations posted each month since iNaturalist was launched in 2008. The number of observations has continued to grow each year breaking 6 million observations per month for the first time this year.

Observations Over Time

This map shows the number of observations by country. Just under half of all iNaturalist observations are from the United States.

Observations by Country

Thank you to everyone who has contributed to the 200 million observations and helped identify them! More than 90 million of these records are shared with the Global Biodiversity Information Facility (GBIF), which has led to more than 5000 publications.

If you want to look back at how we celebrated past observation milestones, you can read about 150 million, 100 million, 50 million, 25 million and 15 million.

Observations and identifications are the social currency of iNaturalist that keeps the community thriving. If you’d also like to financially support the non-profit operations behind-the-scenes, you can make a gift in dozens of currencies.

Donate $5

Thank you for everything that you do to support iNaturalist!

Our Observation of the Week is this undescribed beetle in the tribe Colobotheini, seen in Colombia by @diegoamaya!

Diego Amaya is a botanist who’s currently at the National University of Colombia. Diego’s work is in the systematics, taxonomy and phylogenetics of angiosperms, and focuses on the family Campanulaceae. Last December, while searching for a member of that family, Siphocampylus bogotensis, he came across the beetle you see above.

That day I went with a friend who is interested in insects, so I was more attentive than usual, taking pictures of every insect I saw.

We were on a road that goes down an elevation gradient to link high areas of the western escarpment of Cundinamarca, with low areas in the inter-Andean valleys. This road is known as the Gualivá Reserve and is the most diverse and beautiful point (as far as I am concerned) of my hometown, Facatativá.

Diego posted a photo of the beetle to iNaturalist and originally identified it as a member of the net-winged beetle family (Lycidae), which it resembles, but Colombian entomologist @extasiptera corrected him, identifying it as a long-horned beetle (Cerambycidae) - a completely different family!

iNat beetle identifier extraordinaire @borisb identified it to tribe Colobotheini, and then 15 days ago, noted that it’s likely an undescribed species in or related to the genus Confluentia! I reached out to Boris, and he told me

I have compared this beetle with all described species illustrated on this site. Genera Allocarterica, Batesparna, Carterica, Confluentia, Francisparna, Piriana, and Sparna contain Lycidae mimics exclusively, a few also found within Sangaris. None has a colour pattern that approaches the present one: median area of elytra with dark crossband, instead of pale there. A minor difference in pattern would not indicate a “new” species, because mimics are notoriously variable - a prerequisite for evolution into a mimic.

There are a lot more Longhorn beetles with Lycid colouration in the Neotropics, partly with antennae optically dilated in the basal half by dense fringes of hair, which contribute to Lycid , such as https://www.inaturalist.org/observations/29266366 and chromatic interruption of antennae can also be observed here: https://www.inaturalist.org/observations/14476349 (in this case, probably derived from a non-mimic ancestor with “signal antennae” (white tips / intermediate segments - a nearly 100% tropical phenomenon of unclear relevance).

Diego (above, with a species of Drymonia that is under description) says that he’s loved nature since he was a child and explored the Colombian Andes. He was originally into insects and snakes, but of course ended up studying plants. His current work is under Professor Laura Clavijo, director of the Colombian National Herbarium.

Diego was introduced to iNat in 2017, by one of his professors, who was teaching them about citizen science tools. 

By the time the project was over, I was hooked and didn’t want to leave this great app. For a while, including the pandemic, I used iNaturalist as a tool to help identify observations, and at the same time it helped me to polish my taxonomic skills without leaving home, to the point of becoming a curator in the app. 

Once I was able to return to the field, my passion for photography took over, and iNaturalist became a diary, not only to corroborate IDs, but also to organize my photos and tours. Such has been the impact of iNaturalist in my professional life that it’s brought me several great moments. For example I made the first live records of a moth in more than 100 years. And even uploading observations and contacting expert taxonomists has led me to describing a new species of plant, thanks to these knowledge networks.

(Photo of Diego by Felipe Mesa. Some quotes have been edited for clarity.)

- you can follow Diego on Instagram here!

- @borisb’s work has also led to another beetle Observation of the Week!

- both net-winged beetles and long-horned beetles can be quite showy, check out the most-faved observations of each family here and here!

Our Observation of the Week is this remarkable scene captured by @cumaliyesil in Turkey: a female Goblet-marked Damselfly (Erythromma lindenii) with a male’s dismembered abdomen still attached to her body!

Cumali Yeşil credits the Libellen-Freunde Facebook group with really inspiring his passion for odonates (dragonflies and damselflies), particularly members Hartwig Stobbe and Jörg Teumer. He also heard about iNaturalist there, and joined up.

Being able to learn about all the living creatures from India to America instantly, being in the first place that scientific researchers and expert's identify species, as well as their deep knowledge and interest on the issue has increased my admiration of nature. I love being in nature and being with my family in my spare time.

Before heading to the Beyşehir district last month, Cumali checked out iNaturalist and wanted to observe goblet-marked damselflies, hoping to see how climate change has affected them. 

There was wildlife just by the thicket where people were swimming by the lake, chatting and having fun. It was a life that consisted of feeding, finding shelter, and the struggle for survival. All natural and out  in the open. To observe such wildlife has always affected me deeply.

On that day, as I was making observations, I noticed something strange by my side. I didn’t realize what it was at first, but then I was surprised and appalled as I looked closer.

A female goblet-marked damselfly was flying with a dead body of a male still attached to her! At first I tried to remove the dead body from her but I eventually had to leave her to her fate, which still saddens me. 

So how did this situation happen? Well, damselflies have a complex mating process which involves the male (in this species, males are blue) transferring a packet of sperm from the end of his abdomen to a site near the front of his abdomen. Using the cerci (or claspers) on the end of his abdomen he grabs the female behind her head. The female then arches the tip of her abdomen to pick up the sperm packet, making a heart or wheel shape. When she oviposits (or lays eggs), he either guards her from other males by hovering nearby or will continue to hold her with his cerci.

So what happened here? It’s likely that a predator such as a bird struck the male while he was still attached to the female, taking only his head and thorax and leaving the abdomen. Odonate expert @boudot_jp confirms that this does happen and has a photo of it in his book. I also reached out to @ronan_a, who told me

This gives us an idea of ​​the precision of attachment of the cerci on the female so that a connection is maintained as they move. I don't know if it is linked to an active unlocking system (requiring an action from the male) or to a particular engagement/disengagement movement. 

I've seen similar photos before…but I've never seen it myself in the field. It seems that, for European (and probably Palearctic and Nearctic) damselflies, this is quite specific to the Coenagrionidae family. 

A high school German language teacher, Cumali (above) tells me that the semi-aquatic life history of odonates is what attracted him to them. 

It’s inspired me to raise awareness in the community about the picnics or off-road races in the canyons where I make my observations. To further this, I have decided to to save and create odonate habitats with volunteer students at my school.

Even though my profession is not biology, my interest in biology and the anatomy of odonates continues to grow every day. The biggest contribution to my idea was mostly with the members in iNaturalist and their observations of different living species.

(Some quotes have been edited for clarity and flow.)

- PBS Deep Look has a great damselfly reproduction video.

- check out our profile of odonate identifier @@amila_sumanapala!

That’s right, more caterpillar content!

For the past two springs, a few of us here at iNat - myself, Ken-ichi Ueda, and iNat Board member Cat Chang, as well as some other California naturalists - have had the privilege of joining University of Connecticut professor Dave Wagner and other researchers from the California Academy of Sciences and the Carnegie Museum of Natural History as they searched the deserts of the American southwest for caterpillars. Prof. Wagner is the author of Caterpillars of Eastern North America and for years has been working on a companion volume that covers the western half of the continent.

As Elizabeth Kolbert writes in her profile of Prof. Wagner, he likes to travel with a bunch of helpers and he’s enlisted Laura Gaudette (@gaudettelaura), a collections volunteer with the University of Florida’s McGuire Center for Lepidoptera and Biodiversity, for some of these trips. It was amazing to watch her and all the other researchers stay up late into the night documenting and organizing their finds. 

Laura’s a long-time iNatter, and was gracious enough to sit down with me for a chat and discuss her use of iNat and the work this group is doing. I hope you enjoy!

- if you're not a specialist but want to help get observations in front of those who are, you can use the Identify tool to refine observations!

- and you can also use the Identify page to annotate observations, such as labeling lepidoptera observations as adults, caterpillars, pupae, or eggs. Here's a search for all lepidoptera that are missing a Life Stage annotation.

This is the twenty-fourth entry in an ongoing monthly (or kind of monthly!) series profiling the amazing identifiers of iNaturalist.

Since 2016, Kate Farkas (@k8thegr8) has been adding identifications to iNat observations, to the tune of over 319k verifiable observations identified, putting her in the top three of lepidoptera identifiers of North America! But what’s really impressive is that she focuses not on adult leps, but rather caterpillars, for which there is less information. 

Born in Houston, Texas, Kate’s family was involved with the oil business so her childhood was spent in quite a few places, such as Anchorage, Alaska, and Aberdeen, Scotland. “I've always been interested in the biological sciences,” she tells me, 

doing science summer camps and participating in state science fairs. Biology and critters have always been my passion. I remember playing with a snail that had been in my pocket during a school tornado drill at about age 6, and being impressed that the sky was green when I went to take it outside. My parents always encouraged our family to be in nature, especially camping and hiking.

About fifteen years ago Kate got into nature photography, which piqued her interest in insects, as they’re encountered so frequently in nature. However, she’s into it all, and will photograph anything that moves (as well as plants, fungi, and other non-animals).

I was first introduced to iNaturalist at a moth night in Midlothian that was sponsored by the hospital that I worked for. The moth night was led by Sam Kieschnick (@sambiology), our rockstar urban biologist for Dallas/Fort Worth. I got the app and was browsing people's finds soon after. Through Sam and iNaturalist, I have met so many naturalists in Texas, and many other entomologists around the US and Great Britain. I have been honored to collect and send larvae to entomologists around the country as part of their research (under FDA permits). An unusual phenotype of the Hagen's sphinx larva was described and published because I sent some weird looking caterpillars to @teamontagna for her sphinx moth research. I labeled them incorrectly and everything! 

While she currently works in diagnostic imaging, Kate's other main interest is in trying to complete life cycles of lepidoptera, hence her interest in caterpillars. “There are so many known species of butterflies and moths that have very little known about their larvae and hostplants,” she explains. “I don't have any official training in Lepidoptery but I have a bachelor's degree in general biology. That just means I took every biology class available in college. I've learned so much with the help of John and Jane Balaban (@thebals), Marc Epstein (@marcepstein), and many other iNaturalists.”

When it comes to identifying iNat observations, Kate does it a few times a week, searching for lepidoptera of North America and then using references to help her out. 

I use everything from BugGuide, the Janzen website for Costa Rican lepidoptera, Moth Photographers Group, to the ID guide books by David Wagner, the published fascicles available at the Wedge Entomological Research Foundation under the publications tab, updates in the Lep Society magazines and some specialized butterfly ID books like Caterpillars in the Field and Garden.

I have a particular interest in dagger moths and prominent moths, but anything rare is awesome. Anyone near me when I'm viewing the caterpillars will roll their eyes when I gasp at a rare observation for a Clio tiger moth, or Phaeton primrose sphinx.

I love finding observations that are unknown species. The family and sometimes genus are identifiable, so I can go wandering around and figure out which adults don't have documented larval forms. I have been successful in matching range maps for suspected host plants and caterpillars. One determined iNat user @royaltyler even raised and confirmed the suspected larva for Pygarctia abdominalis and its host plant. I had suspected a group of observations were that species for several years.

If you’re photographing a caterpillar, Kate recommends getting at least three photos: one the top of the caterpillar, one of the side showing all the sets of legs, and a photo of the face (aka head capsule). It’s also really important to note the host plant (or make an observation of the host plant and get it identified on iNat). Even then, she reminds us that not all caterpillars (or adults) can be identified to the species level from photos alone. 

Sometimes, caterpillars can't be identified to the species level. Fairly commonly, caterpillars of similar species can't be told apart unless the caterpillar is reared and documented.  Some entire genus levels of caterpillars look exactly the same, like in the Eupithecia species of pug moths.  If the caterpillars eat all the same foods and are in the same region, we just can't know which it is unless it's raised to adulthood.  The same can be said for adult moths and butterflies. The two species Halysidota tessellaris and Halysidota harrisii can't be told apart as adults, unless they're dissected.  But the caterpillars of those are easy to tell apart!  Sometimes one line, or tuft of hair is all that indicates different species of caterpillars. All of the great identifiers on the website get the ID as close as possible, but it can be frustrating to get a family instead of a species level identification.  

Some of those groups are the underwing moths, tiger moths, tortricid leafroller moths, and some of the flower moths. The pug moths (Eupithecia) are reported to have up to 160 species in North America, but can only be identified through genetic sequencing in a lot of cases. 

Even with those challenges, Kate keeps at it:

If I can get one person to care about insects, all the hours ID-ing will be worth it.  The insect world is vital to the ecosystem, and it seems like all people can do is think of them as pests. Caterpillars are a gateway to caring about insects, as they are generally considered harmless, except for the stinging ones of course. I would love people to appreciate the miraculous and complex world of life around us.

- look for another caterpillar-related post on the iNat Blog in a few days!

- check out past Identifier Profiles of lepidoptera identifiers, @darielsaqui and @michelledelaloye, here and here!

- take a look at our interview with @sambiology, as well as a recent video about his work with iNat from Texas Parks and Wildlife!

If you have someone in mind to be featured in an Identifier Profile, please message @tiwane.

- if you're not a specialist but want to help get observations in front of those who are, you can use the Identify tool to refine observations!

- and you can also use the Identify page to annotate observations, such as labelling lepidoptera observations as adults, caterpillars, pupae, or eggs. Here's a search for all lepidoptera that are missing a Life Stage annotation.

Our Observation of the Week is this trio of Phallus rubrovolvatus stinkhorn mushrooms (アカダマキヌガサタケ in Japanese), seen in Japan by @alinehorikawa!

Born and raised in Brazil, Aline Horikawa and her husband moved to Japan in 2017, and they were walking along Kyoto’s Kamo River last month when they decided to check out a nearby park and botanical garden. 

Walking around and observing so many plants, we noticed a bamboo grove, so we decided to stop and look at what could be there. To our surprise, there were some fungi and we especially noticed the Phallus rubrovolvatus. It was such an incredible surprise that I couldn't believe I was looking at a species like this. I only knew about it from photos on the internet and seeing it in person was wonderful.

Stinkhorn fungal fruiting bodies are pretty fantastical in appearance, with their well, phallic, stalks and lacy veils, but many of us would notice them first by odor. Their spores are contained in a thick, fragrant substance called gleba, and the smell attracts insects. The insects crawl around in the gleba, which sticks to their bodies and later falls off after they’ve moved away, thereby spreading the spores to new areas. 

I reached out to @masaki_hamaguchi, who provided the first Phallus rubrovolvatus identification to this observation, and asked him for more information about this species. He wrote,

Phallus rubrovolvatus has a reddish volva as its name suggests, which is a notable feature of this species. The veil is often shorter and the reticulation coarser than in complex Phallus indusiatus and its allies in Asia.

There may be several other characteristics, but in this observation, the red volva was visible and the form was typical of P. rubrovolvatus, so I judged it to be this species.

Masaki also linked me to this description of the species, which, according to Google Translate, notes that the gleba of this species has a “fruity” smell. 

Aline (above), says 

I have been interested in nature since I was a child in Brazil. I grew up in the countryside, in my grandparents’ house, with lots of plants and animals. I have always loved birds. I also liked playing in places with lots of greenery. I remember that my grandparents’ house had a waterfall and it was one of my favorite places. You could see fish swimming and occasionally I could see wild animals, snakes, lizards and many mammals, and that was impressive for a child.

In 2013 she started birdwatching (“the best choice of my life”), and her interests in nature have broadened well beyond just avians. 

Years later, I graduated as an Environmental Technician. I have done work that I am very proud of, with environmental preservation and fauna surveys in Brazil. In a region, the main job was to make sure that residents could live with wild animals in a more friendly way. Knowledge changes the world, you improve the quality of life of people and animals.

After moving to Japan she continued observing the “wonderful” flora and fauna there, finding new and exciting things every time she goes out. Through a friend, she discovered iNat in 2019. At first she just used it as a place for ID help, but as she got to know iNaturalist better she realized she could do much more with it. 

I started organizing my records to maintain a database for the region where I live, which is in Shiga Prefecture.The way I see the natural world has improved a lot. Knowledge opens doors and helps with environmental preservation.

I have met many researchers and can contribute important photos and information for serious conservation research, especially with birds, reptiles, amphibians, and insects. This makes me very happy and I can only do it because of iNaturalist. All life is important and this allows me to look at an insect or mammal in the same way.

(Photo of Aline by Gerson Horikawa. Some quotes have been lightly edited for clarity.)

- you noticed the fly hanging out by the mushrooms, right?

- Masaki not only loves mushrooms, he’s also a magician! Check out his YouTube channel for videos on both subjects. 

- take a look at the most-faved observations of stinkhorns on iNat!

We hope you enjoy our June News Highlights! If you missed last month’s highlights, you can catch up here. This month, we’re leading with Urban Ecology stories that show how cities are using iNaturalist to reimagine their relationship with wildlife.

Urban Ecology

A. In California, @cesarestien and colleagues from UC Berkeley used iNaturalist to measure the impact of discriminatory practices on urban biodiversity. They published their findings in the Proceedings for the National Academy of Sciences.

In Australia, researchers used iNaturalist data to understand how people are interacting with informal green spaces in urban environments. Another study used iNaturalist data to understand how far people travel to parks across the United States.

In Italy, @deborabarbato and colleagues described how their Siena BiodiverCity project has prompted urban planners to actively incorporate biodiversity and nature into city development.

Species Discoveries

B. Lost species lists can guide efforts to rediscover species.

With the help of iNaturalist data, @chrysoptera and colleagues published a list of 126 lost birds not seen in the last decade (+9 that have been recently found).

You can compare this list with whats been seen on iNaturalist this decade here which includes this spectacular Yellow-crested Helmetshrike found by @cypseloides last December (note that there are some iNat/eBird vs BirdLife taxonomic discrepancies).

Range Extensions

C. In Great Smoky Mountains National Park, @connorrfad documented the first-ever Queen butterfly. Check out this video where @willkuhn from Discover Life in America recounts the discovery.

D. In Canada, @oridgen10's fly observation was among the first records in the country.

E. In Ecuador, @davidweaver’s observation from the Río Palenque Research Center helped confirm a range extension of the Littoral Glassfrog

Invasive Species

F. In the United States, there have been so many stories tracking the Joro spider invasion in June that iNaturalist was mentioned in both Washington Post and the New York Times on the same day. @diego_h’s observation shows a Joro eating a native two-lined spittlebug.

G. In Europe, @mdv used iNaturalist data to understand shifts in the ecological niches of invading mantids outside their native ranges.

H. In Ecuador, @mario_yanez_munoz, @feranolis, @stefannycuadrado, @keyko_cruz, and @lamador used iNaturalist data to understand how Brown anoles are able to invade Guayaquil.

Conservation

iNaturalist is a powerful tool for detecting declines in species (see examples like this monitoring study and this rarity study on Brazilian amphibians and reptiles). Building on this, the following two examples from Canada show communities using iNaturalist to advocate for the protection of habitat to protect rare species:

I. In British Columbia, local groups are using iNaturalist to call for an emergency order to stop an open-pit mine on one of just two known sites of an endangered fern observed here by @eva7.

J. In Alberta, the Save Nose Creek project is using iNaturalist to advocate for the creation of a provincial park in the lands surrounding Nose Creek.

Connectivity

In addition to protecting land, communities are using iNaturalist to improve and restore land. Often this involves increasing connectivity between protected areas to decrease fragmentation. These stories highlight how iNaturalist is being used to increase connectivity by combatting road mortality.

K. In Costa Rica, @carolina_pinto and colleagues are using iNaturalist to identify wildlife crossing mitigation sites.

L. In Canada, researchers used Musk Turtle observations like this one by @pricklybriar to understand road mortality risks as they move between wetlands.

M. In Oregon, the Department of Fish and Wildlife partnered with the Oregon Zoo to launch the Roadkills of Oregon project on iNaturalist to understand where road kills like this gopher snake observed by @peterolsoy are occurring.

iEcology

Scientists are using images contained within iNaturalist images to advance ecology:

N. Across the United States, @moore-evo-eco and colleagues used iNaturalist images to distinguish dragonflies with ornamented wings such as this common whitetail and found that ornamented species are faring better in human-modified landscapes. This research supports sexual selection theories suggesting that signals like wing ornaments can reduce difficulties finding mates in rapidly changing environments.

O. In North America, researchers used iNaturalist images like this one by @louiseinva to compare the interaction between Mexican grass carrying wasps and host flowers across its range.

P. In Italy, researchers used iNaturalist images such as this one by @laura1544 to study spatial distribution of distinct viper phenotypes

AI Naturalist

Q In Germany, @teja6 and colleagues used iNaturalist images to train models to automatically identify individual species from drone imagery of canopies.

Bioblitzes and Events

R. In California, don't miss this story about how the Snapshot Cal Coast 2024 survey led by @rebeccafay and colleagues succeeded in detecting one individual of the critically endangered sunflower sea star that has dramatically declined in recent years.

Also this month, there were many great iNaturalist events for National Pollinator Week. We loved this tutorial by @jdo77 on how to create a pollinator project on iNaturalist.

iNaturalist’s Education Impact

S. Campus bioblitzes like the University of Southern Malaysia Campus Nature Challenge which documented this clouded monitor by @rusidah are a great way to teach conservation and environmental awareness.

There were also many examples of courses and student projects using iNaturalist this month.

iNatters in the News

T. @alan_rockefeller and @mandiequark’s mycology work and use of iNaturalist was featured in the New York Times this month.

U. In Trinadad and Tobago, @akilahstewart published an article on her experience joining the Inter-island Bioblitz

V. In England, @ed-in-sidmouth published an article on how churchyards and cemeteries around Sidmouth can be sanctuaries for nature.

W. In Australia, this story highlighted @joelirwin’s many contributions to the Griffith University Gold Coast Campus bioblitz.

Thank you to everyone who participated on iNaturalist this June! You can become an iNaturalist supporter by clicking the link below:

Donate to iNaturalist

We are excited to announce the launch of our Vision Language Demo, developed in collaboration with our long-time partners at the University of Massachusetts Amherst, the University of Edinburgh, the University College London and MIT, with generous support from Microsoft AI for Earth. This demo enables you to search a snapshot of 10 million iNaturalist photos using text queries. For instance, typing in "a bird eating fruit" will return matching photos ranked by their relevance to your query.

By clicking the “View these observations in Identify” button at the bottom, you can open these photos in the iNaturalist Identify tool where you can add the observations to projects or add observation fields or annotations. We are excited to learn if you find this tool useful for finding and organizing observations representing different life stages (“a caterpillar”), flowering phenology (“a cluster of red berries on a leafy green branch”), captive/cultivated (“a houseplant in a pot”) etc. into projects and with annotations.

Unlike the iNaturalist Computer Vision Model and Geomodel which we train ourselves off of iNaturalist observations, we did not train this model nor is it trained on iNaturalist data. This demo is built off a freely available Vision Language Model that was trained on millions of captioned images not necessarily relating to the natural world. This means that it knows about other things in addition to living organisms (e.g. "a bird perched on a car") but it also means that it currently has biases and may return inappropriate or offensive results that we don’t fully understand. Please keep that in mind when using it.

You can help us and our research collaborators understand how this model (or other Vision Language Models models we may explore or build) perform by clicking on the “Help us Improve” button. By marking the photos on the page that are relevant or not relevant to your search (e.g. "Mating dragonflies") and clicking submit we will be able to compare the performance of different Vision Language Models at this image retrieval task.

We built this demo to better understand the potential of Vision Language Models to help the community organize, explore, and explain the information contained within iNaturalist images. Building this demo has helped us understand the opportunities and challenges associated with this new technology. For example, while these models sometimes demonstrate a surprising ability to describe what is happening in images at a coarse level, they also fail to grasp more complex, finer concepts such as species names.

Two exciting possible future avenues are:

1. Helping to explore and organize iNaturalist images

iNaturalist data have been used in more than 4,900 scientific publications. While many scientific applications stem from qualities of the data that are already easy to filter (species, location, date etc.), an increasing number of studies are leveraging “secondary data” contained within the images themselves ranging from species interactions, to animal behavior, to phenotypic patterns revealed in the images such as color. Here are some examples of recent published studies that resulted from pulling patterns out of iNaturalist images:

Conservation: Recovery plans for the endangered Red-bellied Macaw were premised on the belief that the parrot relies on fruits from a single species of palm for food. Silva and colleagues examined iNaturalist images of this parrot eating fruit and found that it has a much more diverse diet than previously thought.

Climate Change: To reveal how plants are adapting to climate change, Funkano and colleagues examined iNaturalist images of wood sorrels from around the world and found that leaf color is evolving to become redder in urban heat islands.

Animal Behavior: Jagiello and colleagues examined thousands of iNaturalist images of hermit crabs and found that they are increasingly utilizing lighter weight plastic trash in lieu of shells. This study reveals how certain animals are able alter their behavior to take advantage of the Anthropocene and resulting impacts on the ecosystem.

Evolution: Most mammals are thought to have brown eyes. Tabin and Chiasson examined iNaturalist images to test this. They found an exception in the cat branch of the family tree where eye-color is extremely variable and explored the role that sexual selection plays. This paper was covered by Science magazine.

Mimicry: Muñoz-Amezcua and colleagues used computer vision models to examine iNaturalist images and found that many more insects mimic spiders than previously thought. This study reveals how in addition to more efficiently surfacing patterns that the human eye can detect (e.g. cats with blue eyes), vision models can also detect patterns that have gone undetected (e.g. moths that resemble jumping spiders).

We’re very excited to explore whether Vision Language Models can make it explore and organize the rich data contained within iNaturalist images.

2. Explaining Computer Vision species identifications

As anyone using tools such as ChatGPT knows, multimodal Vision Language Models can help explain images in a way that complements more traditional Computer Vision systems. The iNaturalist Computer Vision AI does a great job of telling us what species is in a photo, but it doesn’t do a great job of explaining why that species is suggested.

Offering explanations is something the identifier community does quite well by sharing expertise in text remarks (e.g. “This is Striped Rocket Frog and not Rainforest Rocket Frog because the white stripe extends from the eye to above the leg rather than to the groin.”). We’re interested in building Vision Language Models trained on iNaturalist images and remarks that will help iNautralist users understand why the Computer Vision AI is suggesting certain species and how to distinguish between them.

Deeply integrating Vision Language Models into iNaturalist is still far off and will require new funding opportunities and lots of product and engineering work. But we are very excited to share this small milestone on that journey. Please share your feedback on this exciting new demo!

Our Observation of the Week is this Tatra Chamois (Rupicapra rupicapra tatrica, Kamzík vrchovský tatranský in Slovak), seen in Slovakia by @terana!

“This photo was taken one summer morning in 2013 near the mountain hut where I worked as a student,” says Linda Majdanová, who graduated with a degree in Ecology and Biodiversity Protection. 

When I had free time, I wandered around the surrounding peaks and watched chamois for hours. They are unusually hardy animals. I have many wonderful experiences with them, it is amazing to watch chamois cubs playing together on the remaining snow. Today, meeting a chamois during a hike in the Tatras is not so exceptional, as their numbers, decimated by hunting in the past, have increased thanks to protection. But it is always a pleasant experience to meet this iconic animal, which is undoubtedly a symbol of the Tatras.

A subspecies of the northern chamois, Tatras chamois are herbivorous goat-like mammals which inhabit the Tatras mountains in Slovakia and Poland. As Linda says, they were commonly hunted, with only a few hundred remaining by the end of the 20th century. With conservation work, over a thousand are counted pretty consistently in the mountains now, see the census numbers on its Wikipedia page.

Linda (above) remembers always loving nature, and says 

We lived near the forest and I went on secret trips to nature almost every day and brought home all kinds of bugs, frogs or mushrooms. I was always extremely curious and wanted not only to see, but also to know what I found around me, so instead of fairy tales and children's books, I read educational encyclopedias.

She’s now in the middle of her doctoral studies at the Department of Forest Ecology of the Czech University of Life Sciences in Prague where she specializes in wood-inhabiting fungi in primary and old-growth forests - the topic of her dissertation.

I started contributing to iNaturalist only recently, so I gradually upload older recordings in addition to current observations. I was quite surprised that there are many experts who are able to identify different groups of organisms, which is amazing. Since I spend a lot of my working and free time in the field, whenever I see something interesting or something I don't know, I tend to take a picture of it. And now I know that my observations will not fall into the dust somewhere in a file but will contribute to the knowledge of the world's biota. I also like to look at species distribution maps and find out what lives/grows in different parts of the world.

(Photo of Linda by Vladimír Ruček.)

- this video has some excellent chamois footage.

Each time you post an observation of a plant, additional data can be collected for the study of phenology—the timing of events like flowers, fruits, and leaf color change.

Now, you can annotate all observations of vascular plants to indicate evidence of leaf buds, leaves, seasonal color change, or no leaves.

This is what the options look like on an observation of a vascular plant:

Also, we’ve simplified the annotations related to flowers, fruits, and seeds. The appropriate option(s) should be selected based on the evidence provided in the photos.

If you hover over the options with your cursor, you will see the definition. You can also review the definitions in our documentation.

On taxon pages, by default you see a chart of overall seasonality. Take a look at the Japanese Maple. If you add a place filter in the upper right to restrict it to Japan, you can see that it is most observed in April and November.

If you click on “Flowers and Fruits” and hover over the points on the chart, you can see that April has observations in all four categories: flower buds, flowers, fruits or seeds, and no flowers, fruits or seeds. Note: the chart settings offer the option to “Hide ‘No Annotation’” which was selected here.

If you click on “Leaves” and hover over the points on the chart, you can see that November has many observations with colored leaves.

It is hard to strike the balance of broad applicability and useful specificity for almost 400,000 species of plants. Although many species do not fit nicely into these categories, the goal is to provide a useful starting place for most species of plants. We worked closely with other phenology programs (USA National Phenology Network and Budburst) and phenology researchers to develop and revise these terms.

iNaturalist observations have already been used for many studies of plant phenology, such as the impact of climate change on the wood anemone, spatial and temporal gradients in flowering phenology across Europe, and anomalous Yucca blooms across southwestern North America. Increasingly, researchers want to access more data by using machine learning to label fruits, flowers, and leaves based on a set of training data. With 8 million annotated and verifiable angiosperm observations (out of 71.8 million angiosperm observations), iNaturalist is a massive and growing source of data to understand plant phenology. Now with the ability to add leaf annotations, we can make 76 million vascular plant observations more useful, too.

These updates were supported by a collaborative grant from the US National Science Foundation to advance plant phenology research through the creation of Phenobase. Phenobase will aggregate plant phenology data from many sources using the Plant Phenology Ontology to maximize data interoperability, in addition to using machine learning to infer phenological stages from photos. The Phenobase collaboration involves the USA National Phenology Network, University of Arizona, Louisiana State University, University of Florida, and the Chicago Botanic Garden.

To learn more about adding annotations, you can review the definitions in and a tutorial about using the Identify tool to add annotations quickly using keyboard shortcuts. Now you can see on your profile how many observations you have annotated. Thank you to everyone who annotates observations!