这是indexloc提供的服务,不要输入任何密码

Journal archives for February 2025

February 11, 2025

This Fungus Grows on Horns and Hooves! - Observation of the Week, 2/11/25

Our Observation of the Week is this Horn Stalkball fungus (Onygena equina), seen in Ireland by @paddythecurious!

I think most of us are used to seeing fungi growing on soil, or decomposing logs, but iNatter Paddy Kiely came across a fungus that specializes in a different substrate in southern Ireland, where he lives.

Onygena equina I have come across in two different areas around Tipperary. Both in hilly areas with a forest close by. For this observation, I was looking for wax caps in the field adjacent to the forest. In the forest there was many sheep carcasses and that's where I found the fungi.

Look closely and you’ll see no fungi on the sheep’s bone, just its horns, and that’s because horns are made of keratin (just like fur, claws, reptile scales, and the like), and Onygena equina grows on nothing else. It’s able to break down the keratin in horns and hooves, and then of course it will grow fruiting bodies from which it releases its spores. It’s reputed to sometimes infect human fingernails as well, although I personally haven’t been able to find a peer-reviewed paper showing that - if you know of one, please share in the comments!

Paddy (above, with a Boletus edulis mushroom) says he’s fascinated by “everything fungi.”

I fell down the rabbit hole 7 years ago and learn more and more every day. I have maybe ten books, my favorite, The Fungal Wheels of Europe edition 1 and 2. Also I use a microscope to check spores. I help moderate in 4 groups in Facebook and triage and identify for emergency identification for plants and mushrooms, which I was delighted to be asked to help out in a few months ago. I love to eat mushrooms and have eaten over 150 species in the south of Ireland. I love to find every type of mushroom. Basically I'd like to know about them all. Every day is a treasure hunt.

I use iNaturalist to log my findings of fungi, plants and animals. I believe this is very important for scientific data and research. Also, it's good to look back on for my own research.

(Some quotes have been lightly edited for clarity.)


- other fungi that grow on keratin may be more familiar to humans: ringworm and athlete’s foot. They’re all in the order Onygenales.

- just found out about this fungus yesterday, which grows on coniferous resin. Amazing.

- just a reminder that if you're interested in foraging, one shouldn't rely solely on iNat suggestions when determining the edibility of an organism. It's best to seek out local experts and learn from them.

Posted on February 11, 2025 10:25 PM by tiwane tiwane | 8 comments | Leave a comment

February 15, 2025

A Worm with a Fan! - (Belated) Observation of the Week, 2/4/25

Our (belated) Observation of the Week is this Red-trumpet Calcareous Tubeworm (Serpula columbiana), seen in the United States by @jensseattle!

“I am obsessed with the diversity of marine life here in the Pacific Northwest, where I now live,” says Jen Strongin, who traces her love for marine biology back to her childhood on the Atlantic Ocean.

My family spent a couple of weeks every summer at the edge of Nantucket Sound in South Yarmouth, Massachusetts. I remember getting up early with my mom,  walking the beach and looking for horseshoe crabs and conchs. The beach was (and still is) endlessly fascinating to me with all of its weird and wonderful inhabitants.

A few weeks ago Jen visited a floating dock in Tacoma, Washington, and did her thing:

I was…lying on my belly, hanging over the edge of the dock, looking at all the animals who had decided to settle there. Floating docks are amazing places to observe marine life! There were several Serpula columbiana, looking gorgeous with their cirri all fluffed out. I still cannot get over how beautiful some of our marine worms are here in the Pacific Northwest. I love photographing them.

True to their common name, tubeworms live in a tube they create from calcareous secretions. Much of the worm’s body stays in the tube, but it can extend feathery radioles, or tentacles, which is what Jen has captured here. The radioles are used for both filter feeding and gas exchange. If disturbed, the worm will quickly draw them in and plug the hole to its tube with an operculum.

“I have been working with the Seattle Aquarium's Beach Naturalist Program for the past 8 years, helping to educate volunteers and the public about life on our local shorelines,” says Jen (above). She’s also a professional photographer who specializes on marine life and fungi.

iNat is such an amazing resource. I love being able to learn from others, share observations that may be helpful to scientists/researchers, and to be able to keep track of what I have observed.


- you can check Jen’s photography on her website!

- PBS has a nice video about related Christmas Tree Worms.

Posted on February 15, 2025 12:02 AM by tiwane tiwane | 9 comments | Leave a comment

February 19, 2025

A Rare Glimpse of a Flowering Plant in Kazakhstan - Observation of the Week, 2/19/25

Our Observation of the Week is this flowering Climacoptera lanata amaranth plant, seen in Kazakhstan by @bektemirosmonali!

“I have been interested in nature since childhood. At that time, of course, there were no mobile phones, so we spent most of our time outdoors, as our family was engaged in raising both small and large livestock,” recalls Bektemir Osmonali, Head of the Plant World Cadastre Laboratory "Institute of Botany and Phytointroduction" in Kazakhstan.

I spent most of my summer vacations far from the hustle and bustle of the city, in a place without electricity or other conveniences. For some, this might have felt like a punishment, but for me, it was the most enjoyable and exciting time.

In university he majored in scientific biology and at first couldn’t decide which field to focus on: microbiology, zoology, or botany. “However,” he says, “after my first practical work, I realized that plant research was my true passion.”

Since 2016 he’s delved deeply into plant systematics, earning his PhD, and his research has focused on the Chenopodiaceae family [on iNat, considered part of Amaranthaceae - Tony], as well as other lesser-known families and genera.

One member of the family is Climacoptera lanata, but Bektemir tells me

it is extremely rare to observe, let alone closely examine, the flowering of Climacoptera lanata and related species, as it occurs within a very short period and is usually not seen in large numbers. Nevertheless, I always try to find flowering specimens whenever possible.

Last year, during an expedition to the Mangystau region, I was fortunate to come across a flowering specimen of Climacoptera lanata among other fruiting individuals. Moreover, this particular plant was not only in bloom but also had a well-formed, aesthetically pleasing flower, which I managed to capture in a photograph. Of course, I took numerous pictures, but only a few of them turned out truly beautiful.

In most plants, petals are responsible for the main decorative appeal, but in this family, the stamens are often more visually striking. For example, Climacoptera lanata, which belongs to this family, has beautiful bubble-shaped pink stamens. However, this is not its only distinctive feature—during fruiting, it forms a winged, predominantly pinkish fruit that is also quite attractive, especially in fields where Climacoptera lanata is the dominant species.

Bektemir (above) credits his supervisor Polina Vasilievna Vesselova with encouraging his plant photography. 

Ever since she praised my plant photographs, I have been striving to improve my photography skills and select better angles for my shots…The species of [Chenopodiaceae] are relatively unknown and do not have prominent morphological traits that are immediately noticeable. The true beauty of these plants is revealed through macro photography, as their distinguishing features are incredibly small.

There are many reasons why I use iNaturalist. One of the main ones is the opportunity to publish my work openly, as well as to discuss species identification and their morphological features. iNaturalist has given me the chance to develop creatively and improve my photography skills.

(Photo of Bektemir by Polina Vasilievna Vesselova. Some quotes have been lightly edited for clarity.)


- you can check out Bektemir’s research here, and his YouTube channel here!

- our last Observation of the Week from Kazakhstan was the first known photos of a jumping spider species by @talgar-t64!

Posted on February 19, 2025 04:54 PM by tiwane tiwane | 9 comments | Leave a comment

Better Image Matches: Photo Similarity Update to Computer Vision Suggestions

Have you ever tried to identify a caterpillar and wished that the species suggestions didn’t just show you adult butterflies and moths? Now on the website you will see more relevant photos displayed in ID suggestions.

We’ve updated the Computer Vision Suggestions on the web to display Taxon Photos that are most visually similar to the observation photo. This improvement should make it easier to determine which suggestions are the best match—across all taxa! We're highlighting insects here but it works for plants and fungi, too!

What’s New?

Previously, the Computer Vision system displayed the same taxon photo for a species regardless of the observation. Now, it selects the most visually similar Taxon Photo for each suggestion.

Before this update, the system would show the same default photo whether the observation was a caterpillar or an adult moth.


How Does It Work?

The Computer Vision system now calculates visual similarity between the observation photo and the Taxon Photos for each suggestion.

For example, if you observe a White-lined Sphinx Moth caterpillar, the system will compare your image to Taxon Photos and display the most similar one—in this case, a caterpillar rather than an adult moth.

This improvement depends on having high-quality, diverse Taxon Photos that represent different life stages, sexes, and variations.


What Are Taxon Photos?

Taxon Photos are representative images displayed on each Taxon Page. Each taxon can have up to 12 taxon photos, and these are manually curated by the community to help with identification. You can add or edit Taxon Photos by clicking “Edit Photos” in the Curation menu on any Taxon Page.

  • If you delete a Taxon Photo, the change will be reflected in suggestions immediately.
  • If you add a new Taxon Photo, it may take up to 24 hours to appear in suggestions.

To improve the system, Taxon Photos should include clear, high-quality images that represent the diversity of the taxon (e.g., different life stages, sexes, etc.). Here are some taxon photo guidelines you should follow.


How This Improves Coarser Suggestions

This update also helps clarify why the Computer Vision system suggests a higher-level taxon (e.g., genus) when species within that group look very different.

For example:

  • Suggestions of the beetle Genus Asphaera, which includes highly variable species, will now display a genus-level Taxon Photo that visually resembles the observation—instead of the default Taxon Photo which may look very different. This works because all Taxon Photos of species below the genus are being considered.

Change to Taxon Page Links

For species-level suggestions, clicking the “View” link will now open the Taxon Page with the most visually similar photo as the default. However, this feature does not apply to coarser (non-species) suggestions.


Next Steps

We’d love to hear your feedback on this improvement! We plan to roll out this feature to the iNaturalist mobile apps soon.

Let us know what you think and how this affects your experience with Computer Vision Suggestions.

Posted on February 19, 2025 10:37 PM by loarie loarie | 126 comments | Leave a comment

February 25, 2025

New computer vision model with over 100k taxa

We released v2.20 today which is our first model with over 100,000 taxa! It has 805 new taxa compared to last month's 99,545 taxa and is trained off data exported on January 19, 2025.

We're celebrating this 100k milestone by releasing the iNaturalist Open Range Map dataset.

The graph below shows model accuracy estimates using 1,000 random Research Grade observations in each group not seen during training time. The paired bars below compare average accuracy of model 2.19 with the new model 2.20. Each bar shows the accuracy from Computer Vision alone (dark green) and Computer Vision + Geo (green). Overall the average accuracy of 2.20 is 88.7% (statistically the same as 2.19 at 89.8% - as described here we probably expect ~2% variance all other things being equal among experiments).

Here are the new taxa since the last model:
PlantsMammaliaReptilesOther AnimalsMollusksBirdsProtistsFishesFungiAmphibiansArachnidsInsects

Thank you to everyone in the community who contributed the observations and identifications for all of the newly added species!

Posted on February 25, 2025 07:44 PM by loarie loarie | 11 comments | Leave a comment

Celebrating 100,000 Modeled Taxa with the iNaturalist Open Range Map Dataset

A Major Milestone for Biodiversity Mapping

We passed a major milestone with today’s update to iNaturalist’s Computer Vision and Geomodel—100,000 modeled taxa!

To mark this achievement, we’re excited to make the Geomodel Expected Nearby Maps available in three key ways:

This dataset provides a powerful new tool for researchers, conservationists, and policymakers to better understand and protect biodiversity—a key part of iNaturalist's mission.

What Makes the iNaturalist Open Range Map Dataset Unique?

Understanding where species occur and how their distributions are changing is fundamental to biodiversity conservation. Yet, while satellites provide near-continuous imaging of the Earth, most species distribution datasets remain artisanal—often hand-drawn by experts or based on statistical correlations with indirect environmental variables like temperature rather than direct biodiversity observations.

The iNaturalist Open Range Map Dataset is different:

Built on hundreds of millions of observations from the iNaturalist community.
Improving each month as new observations are added.
A true data-driven complement to traditional range maps, reflecting actual species occurrences.

Mapping Species Richness at Scale

One way to showcase this dataset is by estimating terrestrial vertebrate species richness. The map below was generated by stacking 12,000 individual range maps from the dataset, covering approximately 30% of all terrestrial vertebrate species (12k out of 39k). Areas with higher species richness are shown in red.

Despite known biases towards more 'charismatic' species in this subset of vertebrates, this estimate aligns with established biodiversity patterns, showing peak species richness in the tropics, similar to other global biodiversity assessments (e.g., this study).

Similarly, a map of small-ranged terrestrial vertebrates (species with range sizes smaller than the median for their group) highlights endemism hotspots in regions such as the Andes and Madagascar, corroborating findings from other biodiversity studies (e.g., this study).

A Long Road Ahead—But a Clear Path Forward

The maps above focus on terrestrial vertebrates where 30% of all species are in the model. However, the gaps are much larger for other groups. Across all described species, while iNaturalist has documented ~25%, we currently have enough observations to model only 5% of them.

However, where data exists, our models produce accurate, dynamically updated species distribution maps. This gives us hope: if we can continue growing the iNaturalist community and fill in species and geographic gaps, we can provide the conservation community with the high-quality, dynamically updated species distribution data it urgently needs.

Filling the Gaps—Together

To reach the next milestone—200,000 taxa modeled—we need to expand both taxonomically and geographically.

  • Taxonomic Expansion: Observations of lesser-known species, such as insects, plants, and fungi, are critical for increasing coverage beyond well-documented terrestrial vertebrates.
  • Geographic Expansion: The map below is a copy of the terrestrial vertebrate species richness map from above where H3-4 hexagons with fewer than 100 observations of modeled species are grayed out. These gaps represent missing data, but they also represent an opportunity.

By engaging more observers and identifiers worldwide, we can fill these gaps, refine our models, and build an open, data-driven map of life on Earth.

Join us in making this vision a reality! Click the map to zoom in. Can you help spread the word about iNaturalist to help fill these gaps?

Posted on February 25, 2025 07:44 PM by loarie loarie | 27 comments | Leave a comment

Archives