This is the fifth entry in an ongoing monthly series of blog posts highlighting the amazing identifiers of iNaturalist.

Born in the Neuquén Province of Argentina, Michelle Delaloye tells me “I have been interested in nature since I can remember.”

My parents can tell you how I used to go out to look for frogs in the stream, or how I collected bugs in jars. There is something about nature and its inhabitants that has always fascinated me, which is why I dedicated myself to learn about them. Fortunately, my parents always encouraged me, buying encyclopedias, taking me camping, or tolerating the jars with small animals that I had at home to observe them closely and then release them.

During my teenage years, thanks to a book very dear to me (El Gran Libro de la Naturaleza Argentina), I became fascinated with the landscapes and species of my country. Before that, documentaries and encyclopedias only told about species from other countries and continents, and I hadn't been able to learn much about native nature till that time.

After deciding to study biology, Michelle moved to La Plata, Buenos Aires, as there weren’t opportunities in her home region. She’s wrapping up her undergraduate degree in the Licenciatura en Biología con orientación en Zoología program at Universidad Nacional de La Plata (see her academic page here) and her current interest is “mostly Argentine butterflies; their distribution, diversity, habits, ecology, conservation and biology.” She’s currently the top identifier of butterflies in both Argentina and the continent of South America overall.

Butterflies, though, are a bit of recent interest for Michelle. In 2012 she started birding (see her eBird profile here) and tells me her classmates assumed she was going to study birds. Yet her partner (whom she met through a birding club) gave her the Mariposas de la Ciudad de Buenos Aires y alrededores field guide as a present. “I feel that it was there that my particular interest in that group of arthropods arose,” she says. 

Sometimes during birdwatching trips when someone asks me why I like butterflies, I always comment that they would be the equivalent of birds, but in insects. Not only do they have a diversity of colors and shapes, but they can also have territorial behavior, courtship, migration and other interesting behaviors. They are intimately related to the environment through their nutritional plants, and of course, have metamorphosis, one of the most fascinating processes in the natural world. It should also be mentioned that except for some naturalists such as Núñez Bustos, the study of butterflies has been very underdeveloped in the country for several decades, something that I consider raises my interest in the group, despite the difficulties involved in learning about them with the limited bibliography available.

After seeing a Facebook post for the 2018 City Nature Challenge (La Plata and Buenos Aires were the only Argentinian cities participating that year), Michelle signed up for an iNaturalist account. 

Although I uploaded very few photos to iNat for the CNC that year, I ended up staying for the infinite possibilities it allows. Among them the most interesting thing is to be able to create a database of local biodiversity at different scales, almost like a constantly updated online encyclopedia containing species, their common names, distribution, flowering times, life stages, etc. An absolute wonder.

Now, most of her iNaturalist activity consists of making identifications for other users. “Although I know about plants, birds, and other insects in addition to butterflies,” she explains, “the usual thing I concentrate on is butterflies, which is where I can contribute the most since there are not many users yet who are dedicated to identifying Argentine species.” She tries to add IDs daily (except during exam times) and often starts through her Mariposas y Polillas de Argentina project but will also search for butterflies in other parts of the continent as well as review and correct (if necessary) Research Grade observations. “As I expand my expertise,” she says, “I also expand the map of southern South America a bit more to review.” Sometimes she’ll let observations of difficult genera or tribes accumulate, then she’ll pore over specific literature as she goes through them, and she’s also created notebooks to help her with making identifications (below).

“There is something enjoyable about helping other people name what they observed,” Michelle says, when asked why she identifies observations on iNat.

Even more so when I see, over time, that they learn to recognize many of these species, and begin to add their own identifications. I consider it almost a duty to transmit the knowledge that one possesses to other people, it is something that I always embrace in the birdwatching outings - and as a would-be biologist - and anyone who wants me to explain how I came to the identification, I will always comment on it…

On the other hand, I also learn by identifying the observations of other people: there are entire groups of species that I do not know in person yet, or that are in places that I have not visited yet, and thanks to the observations that other users upload, I can practice recognizing them. In truth we learn both, the observer and the identifier. Finally, thanks to the observations that users upload (and that I and other users identify), not only do I learn more about Argentine species, their distribution, life stages and host plants, but we all learn, and contribute to our knowledge of the country's biodiversity.

(Photo of Michelle taken by Natalie Dudinszky)

- Michelle will be part of a panel discussing insect observations this Wednesday, October 6th, at 6 pm Buenos Aires time. This a series of videos being created by ArgentiNat in the run up to the Great Southern Bioblitz. Check out their blog post for more information and links to recordings of past tutorials.

- When asked what her favorite butterfly taxa are, Michelle replied “It's hard to pick, but I do find interesting the myrmecophilous species, such as Aricoris arenarum, fed and tended by Camponotus ants. Morpho epistrophus is the southernmost Morpho species, reaching as far south as General Madariaga Partido, in Buenos Aires Province, away from any tropical forest (as one would typically expect for a Morpho). Itylos titicaca is one of the smallest butterfly species, with only 10-16 mm of wingspan; it flies in the High Andes, between 3000-4500 m. I find stunning the silver color of Argyrophorus argenteus and the golden one of Argopteron aureipennis. I like Ancyloxypha nitedula, Doxocopa laurentia, Cissia phronius, Heliopetes americanus, Panca subpunctuli and the Butleria and Dardarina species. It's really hard to choose!”

- In addition to her notebooks, Michelle uses both websites and printed guides, as well as a collection of PDFs, checklists, photos, and such. She also uses Butterflies of America’s lists, which she notes are sometimes out of date but are still very useful.

- ArgentiNat is the Argentina node of the iNaturalist network, and it’s managed by Fundacion Vida Silvestre Argentina. You can read more about it here.

Our Observation of the Week is this group of Salpingoeca protozoans on Hydrodictyon algae! Seen in Spain by @vicentefranch.

While most of Vicente Franch’s observations come from southeastern Spain, where he resides, the organisms you see above were found at the Guadalix reservoir in the Sierra de Guadarrama north of Madrid when he was on a family visit. 

The day before it had rained in the Sierra, so the water accumulated in the reservoir at a level a little higher than usual in summer. However, I could get to the edge of the water by wetting my boots.

In the sample I have found a very rich microdiversity. Among them was a precious network of the alga Hydrodictyon used by this group of Salpingoeca, a relatively common sessile coanozoan, as a substrate.

Also known as water nets, algae in the genus Hydrodictyon grow in colonies which often form hexagonal or pentagonal structures. It reproduces both sexually and asexually and can grow quickly, making it a nuisance in some areas. The Salpingoeca protozoans found on this colony are choanoflagellates, which are thought to be the closest unicellular relatives to animals. These prey on bacteria and other biological matter.

Vincente (above, in Valencia lagoon) says he’s always been interested in small animals, which 

awakened in me a certain fascination. My parents bought me a children's microscope with which I spent hours in my room gutting cockroaches and trying to find strange and wonderful things in the dirty waters of the puddles.

This led to him studying biology, but throughout his professional life he wasn't able to interact with nature as much as he wanted to. Now that he’s retired, he’s been able to reconnect with his old hobby, and tells me that “for several years I have been able to improve my observation equipment and spend more time on microscopic life.”

- Here’s a video showing aHydrodictyon colony at various focus settings.

- Go, flageullum, go!

- Let’s go back to April of 2016 and our Observation of the Week blog post featuring @sarka and a totally different algal colony.

Our Observation of the Week is this Bold Jumping Spider (Phidippus audax) and its Spotted Lanternfly (Lycorma delicatula) prey, seen in the United States by @mo0nsgreenthumb!

“I have always been interested in nature and considered myself somewhat of a ‘tree hugger,’” says Rita Tomassetti (@mo0nsgreenthumb), “although I have to confess I used to be terrified of any and all insects, and especially spiders. This all changed last year when I turned my failed efforts at gardening into a success by planting native plants.”

Originally from the Dominican Republic, Rita moved to Pennsylvania in the United States about eight years ago. She wanted to attract butterflies and other pollinators to her garden but found that the pretty non-native plants weren’t doing the job and instead became overrun by weeds. “It was very discouraging until 2 years ago when I attended a native plant sale and bought a Butterfly Garden in a box. It contained 2 beebalms, 1 coreopsis and 1 butterfly milkweed. I planted them in my front garden bed and suddenly I found 3 monarch caterpillars, then found their chrysalis and saw them fly off as butterflies.”

[In 2020] I joined a Facebook group called Gardening for Pollinators and Wildlife Conservation where I learned about all the different kinds of native plants I could plant to attract all the pollinators. I prepped the back garden, went plant shopping and planted densely. Everything did very well and that year I saw my previously sterile garden teeming with life. I learned all about the importance of all the insects visiting my garden, even the scary ones like spiders and hornets and wasps…[it’s] the one positive outcome from Covid and the lockdown.

I am very passionate about native plant gardening and taking care of this amazing Earth we have been blessed with and using iNaturalist I have been able to share with others on my Instagram and Facebook not just the benefit of planting natives but the benefit of having the myriad of insects visit the garden. How we don't need to be afraid of that bee or wasp or fierce looking beetle because they have an important role as pollinators and/or predators, and if we spray them to death we are not just harming them but we are harming ourselves in the long run.

However, the photo you see above was not taken in Rita’s garden (the one below was, though!). She was participating in the Amazing Arthropods 2021 project but

it was a very busy week for me and I barely had time to go out to my garden, but sometime that week I had stopped at a local bakery, on my way to pick up my son from school, and out of the corner of my eye, on a hydrangea leaf, I saw the bold jumping spider and spotted lanternfly so I snapped a quick picture and went on my way. I live in Pennsylvania and as you are all well aware I'm sure, the spotted lanternflies have been a major problem...I had also seen jumping spiders feasting on them before, so I was not surprised to see this spider enjoying its spotted lanternfly meal.

As Rita alluded to, the spotted lanternfly is having a bit of a moment in the United States press. Pieces about this invasive species have popped up in The New York Times, Science Friday, and many local outlets in the northeastern part of the country. Originally from parts of mainland Asia, it was first confirmed in the United States 2014, in Berks County, Pennsylvania but has spread to several other states. Check out the gif @kevinfaccenda posted on our forum. 

A type of planthopper, the spotted lanternfly’s preferred host is tree-of-heaven (Ailanthus altissima) but it will feed on many other plants, including important agricultural ones such as grapes, stone fruit, hardwoods, and hops. When large numbers of them feed on a plant (they suck out its sap), they can damage it by overfeeding or by excreting honeydew, which promotes mold growth that can be lethal the plant. 

Rita’s observation of this bold jumping spider dining on a spotted lanternfly has been added to the Spotted Lanternfly Predation in the U.S. project, where it’s currently one of nearly 100 observations. I reached out to project creator @robizzy, who told me 

Folks were snapping the occasional photo of birds and insects munching on a spotted lanternfly, which was a welcome sight as more and more of Pennsylvania was becoming overrun with them since they arrived a few years back. When lanternflies first showed up in Pennsylvania the fear was that there wouldn't be predators here to eat the invasive insects, so seeing that some things were at least willing to try offered some hope that we had some natural allies. iNaturalist lends itself really well to crowd sourcing these types of observations, so we decided to create a project to document any instances of other organisms taking down lanternflies. 

Around the same time a grad student at Penn State, Anne Johnson, was looking into reports of birds eating lanternflies, so we were happy to contribute this data to her research. The hope is that these projects help us better understand how native, and some non-native, organisms might be able to help us minimize damage from the lanternflies, and help us appreciate allies like our native birds and jumping spiders that much more!

- The US Department of Agriculture has information about spotted lanternflies in the US and what you can about them if you see one there.

- Spotted lanternflies are also an agricultural pest in South Korea, where they were first spotted in 2004.

- I didn’t write much about the the jumping spider part of Rita’s observation, but like all spiders, jumping spiders are awesome! Here are some of the most-faved jumping spider observations on iNat.

Our Observation of the Week is this Massospora diceroproctae-infected Florida Keys' Scrub Cicada (Diceroprocta biconica), seen in the United States by @oridgen10!

During the Brood X mass periodical cicada emergence in North America this year, many cicada watchers noticed a strange affliction on some of the emergent adults - their rear ends were gone and had been replaced by a crusty looking fungal mass. This was likely Massospora cicadina, a parasitic fungus that parasitizes only periodical cicadas. There’s a lot we don’t know about these fungi, but what we do know is pretty amazing. 

It’s thought that Massospora spores infect cicadas when they’re still underground, in their mature nymph stage. This is known as a Stage I infection. After they metamorphose, the fungus spreads into the adult host’s abdomen, eventually making part of it slough off and basically “replacing” it with a mass of spores. Even more amazing, at least some Massospora species alter the behavior of its host via chemicals. For example, some infected male cicadas begin behaving like females, causing amorous males to attempt mating with them and thus contracting the spores. This results in a Stage II infection, and the now-infected adults spread the spores that will fall onto the ground and eventually infect the nymphs. You can read more about some of these behaviors here. 

Not all cicadas, however, are periodical - many are “annual,” with emergences happening every year and not as staggered as the periodical species. And of course, there are Massospora species which host on annual cicadas.

Which brings us to Owen Ridgen, currently a student at the University of Toronto, who credits his dad (who in turn was inspired by his dad) for his interest in nature. “I began as mainly a birder,” he says, “but as I've grown older I've branched out and begun to appreciate all forms of life. I have a special interest in odonata (dragonflies and damselflies), lepidoptera, fishes, reef life, freshwater mussels, and reptiles and amphibians.”

While vacationing in the Florida Keys in August, Owen went mothing - “essentially just setting up a white sheet and reflecting a UV or mercury-vapour bulb off of it in order to attract moths, which can then be photographed” - and noticed that a good number of cicadas were also being drawn to his sheet. 

I began to notice that a significant proportion of these cicadas seemed to be missing the back half of their abdomens. One of them even had some chalky-looking material attached to this area. So I uploaded an observation of this cicada to iNat.

Cicada researcher @willc-t noticed Owen’s observation and recognized that the cicada had a Massospora infection. So he in turn emailed Matt Kasson (@millipedeeats, below), a professor at the University of West Virginia.

Dr. Kasson and his lab have been studying this genus since 2016, and in 2020 his lab published the first phylogenetic study of these fungi. Their Massospora diceroproctae specimen came from a Diceroprocta semicincta cicada collected in Arizona - thousands of kilometers away from the Florida Keys. He tells me

The original description for the Massospora from D. semicincta, M. diceroproctae, also listed D. biconica [which Owen observed] as a potential host but the author questioned if it was really the same fungus given the wide stretch of real estate between the Keys and Arizona. In 2006 one of my former grad students photographed a Massospora-infected Diceroprocta biconica with the same purple-colored fungal plug we see from D. semicincta. But his observations were a decade before my lab even started studying these fungi.

So, Dr. Kasson reached out to Owen. 

When I reached out I was disappointed he had not collected the specimen (why would he) but was ecstatic when he said he was still in the area and could go back to where he made the observations and secure several more if possible. We exchanged messages via iNat and he was able to collect us several specimens.

[Once the specimens arrived], my post-doc Dr. Brian Lovett and my PhD student Angie Macias, both who work with me on this group of fungi, processed the samples and we made both slides for microscopy and extracted DNA for our molecular work. Since Angie had already built a nicely resolved phylogeny and deposited representative sequences into GenBank it was just a matter of BLASTing our sequence date we generated to see the percent sequence similarity with M. diceroproctae.

A cross-country collaboration via iNat confirmed a new host for this fungus! 

Owen (above) says he’s not currently studying anything but has some ideas for future projects. He has been on iNat since 2017 and tells me

I typically use iNat every day of my life. Pretty much anything I see that I can get a good picture of, I will upload. But it's more than that, it is also a community that I love to participate in. iNaturalist has inspired me to be far more attentive to the natural world, and seek out organisms that I knew little about before. It's helped me gain a much more holistic view of my environment overall.

(Photo of Dr. Kasson by Amy Metheny. Photo of Owen by David Ridgen.)

- @willc-t is also part of this cicada rediscovery from last year!

- Entomologist Samuel Ramsey discusses cicada mating and Massospora in this video.

We've started training on a new model, which will be our first model update since July 2021. Here's what you need to know.

It’s bigger

iNaturalist data continues to grow. This time around we went from 38,000 to 47,000 taxa, and from 21 million to 25 million training photos.

We’ve sped up training time again

iNaturalist has new computer vision hardware!. We have two more NVIDIA RTX 8000 GPUs, again granted to iNaturalist by NVIDIA. Based on early experiments, three GPUs seem to train about twice as fast as a single GPU in flat-out training speed. We also have a new computer vision server to house these GPUs, which has 4x the RAM, a hugely faster CPU, and really fast disks (at this scale, reading photos from disk and writing data back to disk is a limiting factor).

This training run is starting with the last checkpoint from the previous training run, rather than starting from the standard ImageNet weights like we did for the previous training run. Basically, this training run gets a head start in understanding what kinds of visual features are important for making iNaturalist suggestions.

We changed a few things about how we generate training data

Hybrid species are no longer included. The previous training run was the first time where we had a significant amount of training images for some avian hybrid taxa (mallard hybrids, for example), and including them as training categories really confused the model. Despite mallard being the most commonly observed species on iNaturalist, the most recent model had a hard time getting mallard suggestions right, struggling to tell the difference between mallards and the various mallard hybrids. Excluding hybrid species from the training set should keep the computer vision model on the task of trying to distinguish between visually distinct taxa.

The number of photos from an individual observation that could be included in the training set has been capped to a max of 5 photos per observation. Previously, in very rare instances, a single observation with hundreds of nearly identical photos could have dominated the training data for a single taxon, potentially causing the model to learn visual features from just those photos, to the detriment of generalizing well to others’ photos.

We changed one thing about how we train the model

Label smoothing is back in our training config. Label smoothing sets the “true” training labels to “softer” values like 0.9 instead of “harder” values like 1.0. Basically, when the model is shown a photo to learn from, we’re now saying “we’re very confident that it’s species X” instead of saying “we’re 100% convinced that it’s species X.” It’s designed to reduce overconfidence in model scores. This is something that we’ve done with some models in the past, but this configuration got lost in the transition for the previous training run.

When will it be ready?

This new model will take a few months to train, and then a few weeks to test / validate before we decide it’s ready for deployment. The increase in training images is slowing us down, but the new hardware is speeding us up. Starting from a checkpoint from our previously trained model also reduces the amount of training work that needs to be done.

Future Work?

The main priority is to get to our stated goal of training 2 models a year.

We're trying to be more transparent about when and how we train new models, so we’ll be working on changes for that. This post is a start, but you can expect a dashboard with more details and charts coming soon.

We have a grant from Amazon to explore ways to improve how we export and train our models, and we’ll be working on that before the end of the year.

Finally, we now have two computer vision systems, one dedicated to production training runs like this one, and another which we can use to run experiments to improve future training runs and explore other machine learning-based features for iNaturalist.

Our Observation of the Week is the first Green-spotted Torrent Frog (Amolops viridimaculatus, 绿点湍蛙 in Chinese (simplified)), seen in China by @jianwang!

A herpetologist at the Museum of Biology Sun Yat-sen University in Guangzhou, China, Jian Wang’s work focuses mostly on snakes, lizards, and anuras (frogs and toads) and he’s interested in both their taxonomy and conservation. Since he started his work in 2014, he’s “participated in the description of nearly 40 new species and the revision of several species groups,” and you can see his publications here. But this frog was seen far to the west, in China’s Yunnan Province. 

“Yunnan Province has the highest biodiversity among all administrative regions in China,” he explains.

Located in the western Yunnan, the ample monsoon rainfall results in a rich diversity of fauna and flora in the Gaoligongshan Region (高黎贡山). Scientists have conducted biodiversity surveys since the 19th century, and plentiful new species/species with restricted distribution have been discovered. With over 700 recorded vertebrates, Gaoligongshan has become one of the places I’ve most yearned to visit…Herpetofauna of the northern, middle and southern parts of Gaoligongshan are distinctly different due to Gaoligongshan’s wide geographical expanse.

In 2015 he was finally able to make it there, which is where he came across the green-spotted torrent frog in its type locality. “[This species] is mainly distributed in the southwest slopes of Gaoligongshan,” says Jian Wang. “While I was investigating in a bone-chilling stream, this fascinating creature occurred, which escaped and allowed me to make only one shot.”

Like other members of its genus, this species does seem to prefer fast-moving cool streams in mountains and hills, and as tadpole it has a suction cup on its belly to help it stick to rocks. It’s also found in parts of India, Vietnam, Myanmar, and possibly Laos. The IUCN lists it as Near Threatened, with major threats being agriculture, logging, and dams or other water management.  

“We made other fascinating discoveries on the same trip, and several new species were gradually described,” says Jian Wang (above). These include the Western Yunnan Music Frog (Nidirana occidentalis, 滇西琴蛙), the Fei’s Horn Toad (Megophrys feii, 费氏角蟾), and the Banna Parachute Gecko (Gekko bannaense, 版纳伞虎)! 

“I love all other groups of animals in the world that that iNat gives me the opportunity to read about,” he says, “[and] I'm glad to help in any way I can in identifying amphibians and reptiles/sharing interesting observations.”

- @qin_huang, who also has also worked at Sun Yat-sen University, is the subject of a previous Observation of the Week post!

- This video shows a different Amolops frog (Amolops larutensis) in a fast-moving Malaysian stream.